tag:blogger.com,1999:blog-66988819180843021792024-03-05T15:39:15.071-08:00East Africa TechometerBlog (est 17Feb2003). Archived a few times since.
As a Current Affairist, I comment on varied topics of interest. I welcome comments and a good debate. You can also find me on twitter/@prasunchat
Care is taken that anything sourced from the internet is referenced by the URL link, where it was found. All credits to where they are due.Unknownnoreply@blogger.comBlogger116125tag:blogger.com,1999:blog-6698881918084302179.post-63977016777676915942023-12-19T08:10:00.001-08:002023-12-24T09:12:50.514-08:00wish tonight's my last In the abyss, a soul resigned,<div>Hope decays, no solace to find.</div><div>On thorns, I rest, dreams lie shattered,</div><div>"wish tonight's my last," says the heart battered.</div><div><br></div><div>Silent screams in the void's cruel embrace,</div><div>Reason abandoned, leaving no trace.</div><div>In agony's theatre, where shadows amass,</div><div>Life's extinguished, swallowed by morass.<div class="separator" style="clear: both; text-align: center;">
<a href="https://blogger.googleusercontent.com/img/a/AVvXsEgxgEWAFZnwDPCNmDqYtSovnfLjLhJb-HS9rD1Ar_oV3Ruykecf_Of2BDoty1Ud7GibHYgt6VmhqKshPE_GSsQ2qY2JUyuejd3pAYggxeICNJTszPi7PVMY8Nwnkm-UZfrt4BFqTT-Ozm3H1o-rVYtNFmmaLvWIYs96buikJq5s3qm8P0tQMm_eHzrDJxo" imageanchor="1" style="margin-left: 1em; margin-right: 1em;">
<img border="0" src="https://blogger.googleusercontent.com/img/a/AVvXsEgxgEWAFZnwDPCNmDqYtSovnfLjLhJb-HS9rD1Ar_oV3Ruykecf_Of2BDoty1Ud7GibHYgt6VmhqKshPE_GSsQ2qY2JUyuejd3pAYggxeICNJTszPi7PVMY8Nwnkm-UZfrt4BFqTT-Ozm3H1o-rVYtNFmmaLvWIYs96buikJq5s3qm8P0tQMm_eHzrDJxo" width="400">
</a>
</div></div>Unknownnoreply@blogger.com0tag:blogger.com,1999:blog-6698881918084302179.post-37842062854976203032023-11-27T04:57:00.001-08:002023-11-27T04:57:24.258-08:00About OSWASOSWAS project in Odisha info:<div><br></div><div>- Name and description: OSWAS stands for Odisha State Workflow Automation System. It is the state government's most prestigious IT initiative to automate the functions at all levels of the administrative hierarchy of government for smooth functioning and timely delivery of government services¹.</div><div><br></div><div>- Status and progress: The new version of OSWAS application with improved features has been put on soft-launch in July 2018 for making the users familiar with it. It was targeted to make the new system Go-Live in September 2018 after necessary customization. As part of the 5T initiative of Government of Odisha, OSWAS is being extended to all Directorates/ HoDs for their integration with respective Administrative Departments in order to speed up the process of decision making².</div><div><br></div><div>- Stakeholders and partners: The stakeholders and partners involved in the implementation of OSWAS are the government departments, agencies, OCAC, IT department, vendors, consultants, and beneficiaries¹².</div><div><br></div><div>- Impact and outcomes: The impact and outcomes of the implementation of OSWAS are:</div><div><br></div><div> - Increase efficiency and effectiveness of the processes</div><div> - Increase employee productivity</div><div> - Efficient management of data</div><div> - Better communication and co-ordination and advancement towards knowledge-led governance</div><div> - Provision of latest IT infrastructure, connectivity, state-of-the-art data centre and security facilities</div><div> - 24x7 secure access, digital signature, online publishing of notices and circulars, SMS and email notification and real time executive dash board</div><div> - In-built Odia plugin for noting and drafting</div><div> - Mobile and cloud ready with web responsive design</div><div> - Robust disaster recovery link with OCAC, State Data Center and National Data Center</div><div> - Principal applications like correspondence management, file management, file processing, record room, knowledge bank, internal messaging, dash board, notice board, MIS reports, audit management and advanced search engine</div><div> - Common applications like Assembly questions, RTI, online telephone directory, vehicle management, tours management, leave management¹².</div><div><br></div><div>- Budget and resources: Not enough info available.</div><div><br></div><div>Source: dt 11/27/2023</div><div>(1) OSWAS | The Odisha Computer Application Centre. https://www.ocac.in/en/services/schemes/oswas.</div><div>(2) Odisha Secretariat Workflow Automation System (OSWAS) put on soft .... https://orissadiary.com/odisha-secretariat-workflow-automation-system-oswas-put-soft-launch-making-users-familiar/.</div><div>(3) Odisha State Wide Network (OSWAN). https://oswan.gov.in/default.asp?GL=1.</div><div>(4) Orissa State Wide Network (OSWAN). https://www.oswan.gov.in/AboutOswan.asp?GL=2.</div>Unknownnoreply@blogger.com0tag:blogger.com,1999:blog-6698881918084302179.post-41770752951980977022023-11-27T04:40:00.001-08:002023-11-27T04:55:23.057-08:00Top priority of Higher Education Department in Odisha in terms of technology <div><b><u>HED PRIORITIES</u></b> </div><div><br></div><div>- Improving quality and equity of selected institutions and enhancing governance of the higher education system</div><div>- Investing in infrastructure and facilities, such as libraries, labs, computers, and internet</div><div>- Implementing e-governance initiatives, such as online modules for UC submission, laptop distribution, pensioners' portal, etc.</div><div>- Promoting online and blended learning, and providing access to digital resources and platforms</div><div><br></div><div>Source: Conversation with Bing, 11/27/2023</div><div>(1) Odisha Higher Education Program for Excellence and Equity. https://projects.worldbank.org/en/projects-operations/project-detail/P160331.</div><div>(2) How Odisha's university paves the way for higher education expansion. https://www.indiatoday.in/education-today/news/story/how-odishas-university-paves-the-way-for-higher-education-expansion-2457148-2023-11-02.</div><div>(3) Home | Higher Education Department. https://dhe.odisha.gov.in/.</div><div>(4) Department of Higher Education (Odisha) - Wikipedia. https://en.wikipedia.org/wiki/Department_of_Higher_Education_%28Odisha%29.</div><div><br></div><div><br></div><div><b><u>Rashtriya Uchchatar Shiksha Abhiyan</u></b></div><div><br></div><div>RUSA stands for Rashtriya Uchchatar Shiksha Abhiyan, which is a central government scheme to improve the access, equity, and quality of higher education in India. It aims to support the state governments in planning and developing their higher education systems through grants and reforms. Odisha joined RUSA in 2013 and has received funds for infrastructure and facilities, e-governance initiatives, online and blended learning, and quality enhancement¹²³.</div><div><br></div><div>Source: Conversation with Bing, 11/27/2023</div><div>(1) Odisha – RUSA. http://rusa.nic.in/odisha/.</div><div>(2) About RUSA | Higher Education Department - Odisha. https://dhe.odisha.gov.in/Schemes-and-Scholarship/RUSA/About-RUSA.</div><div>(3) Overview – RUSA. https://rusa.nic.in/odisha/overview/.</div><div><br></div><div><br></div>Unknownnoreply@blogger.com0tag:blogger.com,1999:blog-6698881918084302179.post-43966302395354738772023-11-15T06:19:00.000-08:002023-11-15T06:19:34.630-08:00Generative AI and LLM - A primer<p><b>1. Background and Evolution of Large Language Model (LLM)</b></p><p>LLM is a Generative model that can input large set of unstructured data and generate large volume of textual output.</p><p><b><u>1.1 A timeline of LLM evolution :-</u></b></p><p>The development of LLM is not new, It went through a gradual evolution since 2002-3: </p><ul style="text-align: left;"><li>2003 - Bag of words : ML to do natural Language Processing NLP</li><li>2008 - TF-IDF : Multi-task Learning</li><li>2013 - Co-occurrence Matrix : Word embeddings</li><li>2013 - Word to Vec/G Love: NLP Neural Nets</li><li>2014 - Seq to Seq Learning</li><li>2015 - Transformer Models , Attention </li></ul>And then comes the explosion of development on LLM:<br /><ul style="text-align: left;"><li>2019 - ELMOs/BERT/XLNet : Pre-trained Models</li><li>Nov 2022 - OpenAIs GPT3.5</li><li>Dec 2022 - Google's MedPaLM</li><li>Feb 2023 - Amazon's Multimodal-CoT</li><li>Feb 2023 - Meta's LLaMA</li><li>Feb 2023 - Microsoft's Kosmos-1</li><li>Mar2023 - Salesforce's einstien GPT</li><li>Mar 2023 - OpenAI's GPT-4</li><li>Mar 2023 - Google's Bard</li><li>Mar 2023- Bloomberg's LLM</li><li>Apr 2023 - Amazon's Bedrock</li></ul><div><b>1.2 Genesis of Transformer Model</b> : (Ref Research Paper : Google's <i>Attention is all you need, 2017</i>)</div><div>[<span style="background-color: white; color: #222222; font-family: Arial, sans-serif; font-size: 13px;">Vaswani, Ashish, et al. "<a href="https://arxiv.org/abs/1706.03762" target="_blank">Attention is all you need</a>." </span><i style="background-color: white; color: #222222; font-family: Arial, sans-serif; font-size: 13px;">Advances in neural information processing systems</i><span style="background-color: white; color: #222222; font-family: Arial, sans-serif; font-size: 13px;"> 30 (2017)]</span></div><div><br /></div><div>Earlier to 2016, Deep learning models were using <u><b>Recurring Neural Network (RNN)</b></u>, or Neural Network deep-learning based. These were not easy to scale, architecture was linear, sequential, computing one output to pass into next input. Google got in transformer blocks, where it essentially modelled non sequentially. example a sentence to be processed word by word, transformer uses Attention to build a relationship to other words in the input sequence as a block. This makes thinking paralelly, scale much faster, revolution in architecture. Volumes of inputs increased manifold from GPT1, 2 and now 3 and 4 where corpus of data to train models kept increasing with billions of data sets. Transformers brought in the key revolution to LLM, in that whie it still implements encoder-decoder architecture, <span style="background-color: white; color: #555555; font-family: "Helvetica Neue", Helvetica, sans-serif; font-size: 15px;">it does not rely on the use of recurrent neural networks.</span>.</div><div><br /></div><div><p style="background: 0px 0px rgb(255, 255, 255); border: 0px; color: #555555; font-family: "Helvetica Neue", Helvetica, sans-serif; font-feature-settings: normal; font-kerning: auto; font-optical-sizing: auto; font-size: 15px; font-stretch: normal; font-variant-alternates: normal; font-variant-east-asian: normal; font-variant-numeric: normal; font-variant-position: normal; font-variation-settings: normal; line-height: 1.5em; margin: 0px 0px 1.2em; outline: 0px; padding: 0px; vertical-align: baseline;">The transformer architecture dispenses of any recurrence and instead relies solely on a <i style="background: 0px 0px; border: 0px; margin: 0px; outline: 0px; padding: 0px; vertical-align: baseline;">self-attention</i> (or intra-attention) mechanism.<span class="Apple-converted-space" style="background: 0px 0px; border: 0px; margin: 0px; outline: 0px; padding: 0px; vertical-align: baseline;"> </span></p><blockquote style="background: 0px 0px rgb(255, 255, 255); border: 0px; color: #444444; font-family: "Helvetica Neue", Helvetica, sans-serif; font-size: 15px; margin: 0px; outline: 0px; padding: 10px 20px 10px 50px; position: relative; quotes: none; vertical-align: baseline;"><p style="background: 0px 0px; border: 0px; color: #555555; font-feature-settings: normal; font-kerning: auto; font-optical-sizing: auto; font-stretch: normal; font-style: italic; font-variant-alternates: normal; font-variant-east-asian: normal; font-variant-numeric: normal; font-variant-position: normal; font-variation-settings: normal; line-height: 1.5em; margin: 0px 0px 1.2em; outline: 0px; padding: 0px; vertical-align: baseline;"><i style="background: 0px 0px; border: 0px; margin: 0px; outline: 0px; padding: 0px; vertical-align: baseline;">In terms of computational complexity, self-attention layers are faster than recurrent layers when the sequence length n is smaller than the representation dimensionality d …</i></p><p style="background: 0px 0px; border: 0px; color: #555555; font-feature-settings: normal; font-kerning: auto; font-optical-sizing: auto; font-stretch: normal; font-style: italic; font-variant-alternates: normal; font-variant-east-asian: normal; font-variant-numeric: normal; font-variant-position: normal; font-variation-settings: normal; line-height: 1.5em; margin: 0px 0px 1.2em; outline: 0px; padding: 0px; text-align: right; vertical-align: baseline;">– <a href="https://www.amazon.com/Advanced-Deep-Learning-Python-next-generation/dp/178995617X" style="background: 0px 0px; border: 0px; color: #428bca; margin: 0px; outline: 0px; padding: 0px; text-decoration-line: none; vertical-align: baseline;">Advanced Deep Learning with Python</a>, 2019.</p></blockquote></div><div><i style="background: 0px 0px rgb(255, 255, 255); border: 0px; color: #555555; font-family: "Helvetica Neue", Helvetica, sans-serif; font-size: 15px; margin: 0px; outline: 0px; padding: 0px; vertical-align: baseline;">Transformers can capture global/long range dependencies between input and output, support parallel processing, require minimal inductive biases (prior knowledge), demonstrate scalability to large sequences and datasets, and allow domain-agnostic processing of multiple modalities (text, images, speech) using similar processing blocks.</i></div><div><br /></div><div><br /></div><div><b><u>1.3 Three basic sort of LLMs</u></b> (as per "Attention is all you need" paper*):-</div><div><br /></div><div><p style="background: 0px 0px rgb(255, 255, 255); border: 0px; color: #555555; font-family: "Helvetica Neue", Helvetica, sans-serif; font-feature-settings: normal; font-kerning: auto; font-optical-sizing: auto; font-size: 15px; font-stretch: normal; font-variant-alternates: normal; font-variant-east-asian: normal; font-variant-numeric: normal; font-variant-position: normal; font-variation-settings: normal; line-height: 1.5em; margin: 0px 0px 1.2em; outline: 0px; padding: 0px; vertical-align: baseline;">The encoder-decoder architecture has been extensively applied to sequence-to-sequence (seq2seq) tasks for language processing. Examples of such tasks within the domain of language processing include machine translation and image captioning. <span class="Apple-converted-space" style="background: 0px 0px; border: 0px; margin: 0px; outline: 0px; padding: 0px; vertical-align: baseline;"> </span></p><blockquote style="background: 0px 0px rgb(255, 255, 255); border: 0px; color: #444444; font-family: "Helvetica Neue", Helvetica, sans-serif; font-size: 15px; margin: 0px; outline: 0px; padding: 10px 20px 10px 50px; position: relative; quotes: none; vertical-align: baseline;"><p style="background: 0px 0px; border: 0px; color: #555555; font-feature-settings: normal; font-kerning: auto; font-optical-sizing: auto; font-stretch: normal; font-style: italic; font-variant-alternates: normal; font-variant-east-asian: normal; font-variant-numeric: normal; font-variant-position: normal; font-variation-settings: normal; line-height: 1.5em; margin: 0px 0px 1.2em; outline: 0px; padding: 0px; vertical-align: baseline;"><i style="background: 0px 0px; border: 0px; margin: 0px; outline: 0px; padding: 0px; vertical-align: baseline;">The earliest use of attention was as part of RNN based encoder-decoder framework to encode long input sentences [Bahdanau et al. 2015]. Consequently, attention has been most widely used with this architecture.</i></p><p style="background: 0px 0px; border: 0px; color: #555555; font-feature-settings: normal; font-kerning: auto; font-optical-sizing: auto; font-stretch: normal; font-style: italic; font-variant-alternates: normal; font-variant-east-asian: normal; font-variant-numeric: normal; font-variant-position: normal; font-variation-settings: normal; line-height: 1.5em; margin: 0px 0px 1.2em; outline: 0px; padding: 0px; text-align: right; vertical-align: baseline;">– <a href="https://arxiv.org/abs/1904.02874?utm_source=feedburner&utm_medium=feed&utm_campaign=Feed%253A+arxiv%252FQSXk+%2528ExcitingAds%2521+cs+updates+on+arXiv.org%2529" style="background: 0px 0px; border: 0px; color: #428bca; margin: 0px; outline: 0px; padding: 0px; text-decoration-line: none; vertical-align: baseline;" target="_blank">An Attentive Survey of Attention Models</a>, 2021.</p></blockquote></div><div>> Encoder-Only</div><div>> Decoder-only</div><div>> Encoder-Decoder.</div><div><br /></div><div>Let's see what these are :</div><div><u style="font-weight: bold;"><br /></u></div><div><u style="font-weight: bold;"><div class="separator" style="clear: both; text-align: center;"><a href="https://blogger.googleusercontent.com/img/a/AVvXsEj9Jb4W_wPPXZW9RW6Ejde1egOIYpTe6GCTHdvKJkarHjIJkyYuv2Dc20Q74q8kGGlhgUTPrwkxQMkxSzhfKtcOtgyw-s-WuBQKCLrY8ky4c5_p4VFbLDp8OxD3y8Nt6KI4rWrTant8UjZZ49YdNQGiWoI_z0D3ziMnd2mPTSEe-7eyuM-C09cLaSu0dkA" style="margin-left: 1em; margin-right: 1em;"><img alt="" data-original-height="782" data-original-width="861" height="364" src="https://blogger.googleusercontent.com/img/a/AVvXsEj9Jb4W_wPPXZW9RW6Ejde1egOIYpTe6GCTHdvKJkarHjIJkyYuv2Dc20Q74q8kGGlhgUTPrwkxQMkxSzhfKtcOtgyw-s-WuBQKCLrY8ky4c5_p4VFbLDp8OxD3y8Nt6KI4rWrTant8UjZZ49YdNQGiWoI_z0D3ziMnd2mPTSEe-7eyuM-C09cLaSu0dkA=w400-h364" width="400" /></a></div><br /><br /></u></div><div><span style="font-weight: bold;"> <u> 1.3.1 Encoder Only: </u></span><span style="font-weight: bold;"> </span><b>Ex : GPT/OpenAI (content in same language)</b></div><div><span style="font-weight: bold;"><span> </span><span> </span>Compacts/encodes one set of input into something like sentiment analysis</span></div><div><br /></div><div> - Popularized via successful architectures like BERT*</div><div> - Very good for predictive use on unstructured data</div><div><u style="font-weight: bold;"><br /></u></div><div><span style="background-color: white; color: #404040; font-family: "SF Pro Display", -apple-system, BlinkMacSystemFont, Inter, "Segoe UI", Roboto, Helvetica, Arial, sans-serif, "Apple Color Emoji", "Segoe UI Emoji", "Segoe UI Symbol";">Encoder-only models are still very useful for training predictive models based on text embeddings versus generating texts.</span></div><div><u style="font-size: x-small; font-weight: bold;">*</u><u style="font-weight: bold;">[</u><span style="--tw-border-spacing-x: 0; --tw-border-spacing-y: 0; --tw-ring-color: rgb(59 130 246 / 0.5); --tw-ring-offset-color: #fff; --tw-ring-offset-shadow: 0 0 #0000; --tw-ring-offset-width: 0px; --tw-ring-shadow: 0 0 #0000; --tw-rotate: 0; --tw-scale-x: 1; --tw-scale-y: 1; --tw-scroll-snap-strictness: proximity; --tw-shadow-colored: 0 0 #0000; --tw-shadow: 0 0 #0000; --tw-skew-x: 0; --tw-skew-y: 0; --tw-translate-x: 0; --tw-translate-y: 0; color: #404040; font-family: "SF Pro Display", -apple-system, BlinkMacSystemFont, Inter, "Segoe UI", Roboto, Helvetica, Arial, sans-serif, "Apple Color Emoji", "Segoe UI Emoji", "Segoe UI Symbol";">Over the years, various encoder-only architectures have been developed based on the encoder module of the original transformer model outlined above. Notable examples include BERT (</span><em style="--tw-border-spacing-x: 0; --tw-border-spacing-y: 0; --tw-ring-color: rgb(59 130 246 / 0.5); --tw-ring-offset-color: #fff; --tw-ring-offset-shadow: 0 0 #0000; --tw-ring-offset-width: 0px; --tw-ring-shadow: 0 0 #0000; --tw-rotate: 0; --tw-scale-x: 1; --tw-scale-y: 1; --tw-scroll-snap-strictness: proximity; --tw-shadow-colored: 0 0 #0000; --tw-shadow: 0 0 #0000; --tw-skew-x: 0; --tw-skew-y: 0; --tw-translate-x: 0; --tw-translate-y: 0; color: #404040; font-family: "SF Pro Display", -apple-system, BlinkMacSystemFont, Inter, "Segoe UI", Roboto, Helvetica, Arial, sans-serif, "Apple Color Emoji", "Segoe UI Emoji", "Segoe UI Symbol";"><a href="https://arxiv.org/abs/1810.04805" rel="" style="--tw-border-spacing-x: 0; --tw-border-spacing-y: 0; --tw-ring-color: rgb(59 130 246 / 0.5); --tw-ring-offset-color: #fff; --tw-ring-offset-shadow: 0 0 #0000; --tw-ring-offset-width: 0px; --tw-ring-shadow: 0 0 #0000; --tw-rotate: 0; --tw-scale-x: 1; --tw-scale-y: 1; --tw-scroll-snap-strictness: proximity; --tw-shadow-colored: 0 0 #0000; --tw-shadow: 0 0 #0000; --tw-skew-x: 0; --tw-skew-y: 0; --tw-translate-x: 0; --tw-translate-y: 0;">Pre-training of Deep Bidirectional Transformers for Language Understanding, 2018</a></em><span style="--tw-border-spacing-x: 0; --tw-border-spacing-y: 0; --tw-ring-color: rgb(59 130 246 / 0.5); --tw-ring-offset-color: #fff; --tw-ring-offset-shadow: 0 0 #0000; --tw-ring-offset-width: 0px; --tw-ring-shadow: 0 0 #0000; --tw-rotate: 0; --tw-scale-x: 1; --tw-scale-y: 1; --tw-scroll-snap-strictness: proximity; --tw-shadow-colored: 0 0 #0000; --tw-shadow: 0 0 #0000; --tw-skew-x: 0; --tw-skew-y: 0; --tw-translate-x: 0; --tw-translate-y: 0; color: #404040; font-family: "SF Pro Display", -apple-system, BlinkMacSystemFont, Inter, "Segoe UI", Roboto, Helvetica, Arial, sans-serif, "Apple Color Emoji", "Segoe UI Emoji", "Segoe UI Symbol";">) and RoBERTa (</span><em style="--tw-border-spacing-x: 0; --tw-border-spacing-y: 0; --tw-ring-color: rgb(59 130 246 / 0.5); --tw-ring-offset-color: #fff; --tw-ring-offset-shadow: 0 0 #0000; --tw-ring-offset-width: 0px; --tw-ring-shadow: 0 0 #0000; --tw-rotate: 0; --tw-scale-x: 1; --tw-scale-y: 1; --tw-scroll-snap-strictness: proximity; --tw-shadow-colored: 0 0 #0000; --tw-shadow: 0 0 #0000; --tw-skew-x: 0; --tw-skew-y: 0; --tw-translate-x: 0; --tw-translate-y: 0; color: #404040; font-family: "SF Pro Display", -apple-system, BlinkMacSystemFont, Inter, "Segoe UI", Roboto, Helvetica, Arial, sans-serif, "Apple Color Emoji", "Segoe UI Emoji", "Segoe UI Symbol";"><a href="https://arxiv.org/abs/1907.11692" rel="" style="--tw-border-spacing-x: 0; --tw-border-spacing-y: 0; --tw-ring-color: rgb(59 130 246 / 0.5); --tw-ring-offset-color: #fff; --tw-ring-offset-shadow: 0 0 #0000; --tw-ring-offset-width: 0px; --tw-ring-shadow: 0 0 #0000; --tw-rotate: 0; --tw-scale-x: 1; --tw-scale-y: 1; --tw-scroll-snap-strictness: proximity; --tw-shadow-colored: 0 0 #0000; --tw-shadow: 0 0 #0000; --tw-skew-x: 0; --tw-skew-y: 0; --tw-translate-x: 0; --tw-translate-y: 0;">A Robustly Optimized BERT Pretraining Approach, 2018</a></em><span style="--tw-border-spacing-x: 0; --tw-border-spacing-y: 0; --tw-ring-color: rgb(59 130 246 / 0.5); --tw-ring-offset-color: #fff; --tw-ring-offset-shadow: 0 0 #0000; --tw-ring-offset-width: 0px; --tw-ring-shadow: 0 0 #0000; --tw-rotate: 0; --tw-scale-x: 1; --tw-scale-y: 1; --tw-scroll-snap-strictness: proximity; --tw-shadow-colored: 0 0 #0000; --tw-shadow: 0 0 #0000; --tw-skew-x: 0; --tw-skew-y: 0; --tw-translate-x: 0; --tw-translate-y: 0; color: #404040; font-family: "SF Pro Display", -apple-system, BlinkMacSystemFont, Inter, "Segoe UI", Roboto, Helvetica, Arial, sans-serif, "Apple Color Emoji", "Segoe UI Emoji", "Segoe UI Symbol";">). </span></div><div><u><span style="--tw-border-spacing-x: 0; --tw-border-spacing-y: 0; --tw-ring-color: rgb(59 130 246 / 0.5); --tw-ring-offset-color: #fff; --tw-ring-offset-shadow: 0 0 #0000; --tw-ring-offset-width: 0px; --tw-ring-shadow: 0 0 #0000; --tw-rotate: 0; --tw-scale-x: 1; --tw-scale-y: 1; --tw-scroll-snap-strictness: proximity; --tw-shadow-colored: 0 0 #0000; --tw-shadow: 0 0 #0000; --tw-skew-x: 0; --tw-skew-y: 0; --tw-translate-x: 0; --tw-translate-y: 0; background-color: white; color: #404040; font-family: "SF Pro Display", -apple-system, BlinkMacSystemFont, Inter, "Segoe UI", Roboto, Helvetica, Arial, sans-serif, "Apple Color Emoji", "Segoe UI Emoji", "Segoe UI Symbol";">BERT (</span><strong style="--tw-border-spacing-x: 0; --tw-border-spacing-y: 0; --tw-ring-color: rgb(59 130 246 / 0.5); --tw-ring-offset-color: #fff; --tw-ring-offset-shadow: 0 0 #0000; --tw-ring-offset-width: 0px; --tw-ring-shadow: 0 0 #0000; --tw-rotate: 0; --tw-scale-x: 1; --tw-scale-y: 1; --tw-scroll-snap-strictness: proximity; --tw-shadow-colored: 0 0 #0000; --tw-shadow: 0 0 #0000; --tw-skew-x: 0; --tw-skew-y: 0; --tw-translate-x: 0; --tw-translate-y: 0; background-color: white; color: #404040; font-family: "SF Pro Display", -apple-system, BlinkMacSystemFont, Inter, "Segoe UI", Roboto, Helvetica, Arial, sans-serif, "Apple Color Emoji", "Segoe UI Emoji", "Segoe UI Symbol";">B</strong><span style="--tw-border-spacing-x: 0; --tw-border-spacing-y: 0; --tw-ring-color: rgb(59 130 246 / 0.5); --tw-ring-offset-color: #fff; --tw-ring-offset-shadow: 0 0 #0000; --tw-ring-offset-width: 0px; --tw-ring-shadow: 0 0 #0000; --tw-rotate: 0; --tw-scale-x: 1; --tw-scale-y: 1; --tw-scroll-snap-strictness: proximity; --tw-shadow-colored: 0 0 #0000; --tw-shadow: 0 0 #0000; --tw-skew-x: 0; --tw-skew-y: 0; --tw-translate-x: 0; --tw-translate-y: 0; background-color: white; color: #404040; font-family: "SF Pro Display", -apple-system, BlinkMacSystemFont, Inter, "Segoe UI", Roboto, Helvetica, Arial, sans-serif, "Apple Color Emoji", "Segoe UI Emoji", "Segoe UI Symbol";">idirectional </span><strong style="--tw-border-spacing-x: 0; --tw-border-spacing-y: 0; --tw-ring-color: rgb(59 130 246 / 0.5); --tw-ring-offset-color: #fff; --tw-ring-offset-shadow: 0 0 #0000; --tw-ring-offset-width: 0px; --tw-ring-shadow: 0 0 #0000; --tw-rotate: 0; --tw-scale-x: 1; --tw-scale-y: 1; --tw-scroll-snap-strictness: proximity; --tw-shadow-colored: 0 0 #0000; --tw-shadow: 0 0 #0000; --tw-skew-x: 0; --tw-skew-y: 0; --tw-translate-x: 0; --tw-translate-y: 0; background-color: white; color: #404040; font-family: "SF Pro Display", -apple-system, BlinkMacSystemFont, Inter, "Segoe UI", Roboto, Helvetica, Arial, sans-serif, "Apple Color Emoji", "Segoe UI Emoji", "Segoe UI Symbol";">E</strong><span style="--tw-border-spacing-x: 0; --tw-border-spacing-y: 0; --tw-ring-color: rgb(59 130 246 / 0.5); --tw-ring-offset-color: #fff; --tw-ring-offset-shadow: 0 0 #0000; --tw-ring-offset-width: 0px; --tw-ring-shadow: 0 0 #0000; --tw-rotate: 0; --tw-scale-x: 1; --tw-scale-y: 1; --tw-scroll-snap-strictness: proximity; --tw-shadow-colored: 0 0 #0000; --tw-shadow: 0 0 #0000; --tw-skew-x: 0; --tw-skew-y: 0; --tw-translate-x: 0; --tw-translate-y: 0; background-color: white; color: #404040; font-family: "SF Pro Display", -apple-system, BlinkMacSystemFont, Inter, "Segoe UI", Roboto, Helvetica, Arial, sans-serif, "Apple Color Emoji", "Segoe UI Emoji", "Segoe UI Symbol";">ncoder </span><strong style="--tw-border-spacing-x: 0; --tw-border-spacing-y: 0; --tw-ring-color: rgb(59 130 246 / 0.5); --tw-ring-offset-color: #fff; --tw-ring-offset-shadow: 0 0 #0000; --tw-ring-offset-width: 0px; --tw-ring-shadow: 0 0 #0000; --tw-rotate: 0; --tw-scale-x: 1; --tw-scale-y: 1; --tw-scroll-snap-strictness: proximity; --tw-shadow-colored: 0 0 #0000; --tw-shadow: 0 0 #0000; --tw-skew-x: 0; --tw-skew-y: 0; --tw-translate-x: 0; --tw-translate-y: 0; background-color: white; color: #404040; font-family: "SF Pro Display", -apple-system, BlinkMacSystemFont, Inter, "Segoe UI", Roboto, Helvetica, Arial, sans-serif, "Apple Color Emoji", "Segoe UI Emoji", "Segoe UI Symbol";">R</strong><span style="--tw-border-spacing-x: 0; --tw-border-spacing-y: 0; --tw-ring-color: rgb(59 130 246 / 0.5); --tw-ring-offset-color: #fff; --tw-ring-offset-shadow: 0 0 #0000; --tw-ring-offset-width: 0px; --tw-ring-shadow: 0 0 #0000; --tw-rotate: 0; --tw-scale-x: 1; --tw-scale-y: 1; --tw-scroll-snap-strictness: proximity; --tw-shadow-colored: 0 0 #0000; --tw-shadow: 0 0 #0000; --tw-skew-x: 0; --tw-skew-y: 0; --tw-translate-x: 0; --tw-translate-y: 0; background-color: white; color: #404040; font-family: "SF Pro Display", -apple-system, BlinkMacSystemFont, Inter, "Segoe UI", Roboto, Helvetica, Arial, sans-serif, "Apple Color Emoji", "Segoe UI Emoji", "Segoe UI Symbol";">epresentations from </span><strong style="--tw-border-spacing-x: 0; --tw-border-spacing-y: 0; --tw-ring-color: rgb(59 130 246 / 0.5); --tw-ring-offset-color: #fff; --tw-ring-offset-shadow: 0 0 #0000; --tw-ring-offset-width: 0px; --tw-ring-shadow: 0 0 #0000; --tw-rotate: 0; --tw-scale-x: 1; --tw-scale-y: 1; --tw-scroll-snap-strictness: proximity; --tw-shadow-colored: 0 0 #0000; --tw-shadow: 0 0 #0000; --tw-skew-x: 0; --tw-skew-y: 0; --tw-translate-x: 0; --tw-translate-y: 0; background-color: white; color: #404040; font-family: "SF Pro Display", -apple-system, BlinkMacSystemFont, Inter, "Segoe UI", Roboto, Helvetica, Arial, sans-serif, "Apple Color Emoji", "Segoe UI Emoji", "Segoe UI Symbol";">T</strong></u><span style="--tw-border-spacing-x: 0; --tw-border-spacing-y: 0; --tw-ring-color: rgb(59 130 246 / 0.5); --tw-ring-offset-color: #fff; --tw-ring-offset-shadow: 0 0 #0000; --tw-ring-offset-width: 0px; --tw-ring-shadow: 0 0 #0000; --tw-rotate: 0; --tw-scale-x: 1; --tw-scale-y: 1; --tw-scroll-snap-strictness: proximity; --tw-shadow-colored: 0 0 #0000; --tw-shadow: 0 0 #0000; --tw-skew-x: 0; --tw-skew-y: 0; --tw-translate-x: 0; --tw-translate-y: 0; background-color: white; color: #404040; font-family: "SF Pro Display", -apple-system, BlinkMacSystemFont, Inter, "Segoe UI", Roboto, Helvetica, Arial, sans-serif, "Apple Color Emoji", "Segoe UI Emoji", "Segoe UI Symbol";"><u>ransformers)</u> is an encoder-only architecture based on the Transformer's encoder module. The BERT model is pretrained on a large text corpus using masked language modeling and next-sentence prediction tasks.]</span></div><div><u style="font-weight: bold;"><br /></u></div><div><span style="font-weight: bold;"> <u>1.3.2 Decoder Only - </u></span><span style="font-weight: bold;">Ex: BERT Architecture : Lot of text data and need one particular <span> </span><span> </span><span> </span><span> </span>Output like sentiment or topic of a discussion </span></div><div> - Popularised via original GPT models</div><div> - Driving the Gen AI market buzz</div><div><u style="font-weight: bold;"><br /></u></div><div><div><span style="--tw-border-spacing-x: 0; --tw-border-spacing-y: 0; --tw-ring-color: rgb(59 130 246 / 0.5); --tw-ring-offset-color: #fff; --tw-ring-offset-shadow: 0 0 #0000; --tw-ring-offset-width: 0px; --tw-ring-shadow: 0 0 #0000; --tw-rotate: 0; --tw-scale-x: 1; --tw-scale-y: 1; --tw-scroll-snap-strictness: proximity; --tw-shadow-colored: 0 0 #0000; --tw-shadow: 0 0 #0000; --tw-skew-x: 0; --tw-skew-y: 0; --tw-translate-x: 0; --tw-translate-y: 0; background-color: white; color: #404040; font-family: "SF Pro Display", -apple-system, BlinkMacSystemFont, Inter, "Segoe UI", Roboto, Helvetica, Arial, sans-serif, "Apple Color Emoji", "Segoe UI Emoji", "Segoe UI Symbol";">Decoder-only models are used for generative tasks including Q&A. </span></div><div><span style="--tw-border-spacing-x: 0; --tw-border-spacing-y: 0; --tw-ring-color: rgb(59 130 246 / 0.5); --tw-ring-offset-color: #fff; --tw-ring-offset-shadow: 0 0 #0000; --tw-ring-offset-width: 0px; --tw-ring-shadow: 0 0 #0000; --tw-rotate: 0; --tw-scale-x: 1; --tw-scale-y: 1; --tw-scroll-snap-strictness: proximity; --tw-shadow-colored: 0 0 #0000; --tw-shadow: 0 0 #0000; --tw-skew-x: 0; --tw-skew-y: 0; --tw-translate-x: 0; --tw-translate-y: 0; background-color: white; color: #404040; font-family: "SF Pro Display", -apple-system, BlinkMacSystemFont, Inter, "Segoe UI", Roboto, Helvetica, Arial, sans-serif, "Apple Color Emoji", "Segoe UI Emoji", "Segoe UI Symbol";">The <u>GPT (</u></span><u><strong style="--tw-border-spacing-x: 0; --tw-border-spacing-y: 0; --tw-ring-color: rgb(59 130 246 / 0.5); --tw-ring-offset-color: #fff; --tw-ring-offset-shadow: 0 0 #0000; --tw-ring-offset-width: 0px; --tw-ring-shadow: 0 0 #0000; --tw-rotate: 0; --tw-scale-x: 1; --tw-scale-y: 1; --tw-scroll-snap-strictness: proximity; --tw-shadow-colored: 0 0 #0000; --tw-shadow: 0 0 #0000; --tw-skew-x: 0; --tw-skew-y: 0; --tw-translate-x: 0; --tw-translate-y: 0; background-color: white; color: #404040; font-family: "SF Pro Display", -apple-system, BlinkMacSystemFont, Inter, "Segoe UI", Roboto, Helvetica, Arial, sans-serif, "Apple Color Emoji", "Segoe UI Emoji", "Segoe UI Symbol";">G</strong><span style="--tw-border-spacing-x: 0; --tw-border-spacing-y: 0; --tw-ring-color: rgb(59 130 246 / 0.5); --tw-ring-offset-color: #fff; --tw-ring-offset-shadow: 0 0 #0000; --tw-ring-offset-width: 0px; --tw-ring-shadow: 0 0 #0000; --tw-rotate: 0; --tw-scale-x: 1; --tw-scale-y: 1; --tw-scroll-snap-strictness: proximity; --tw-shadow-colored: 0 0 #0000; --tw-shadow: 0 0 #0000; --tw-skew-x: 0; --tw-skew-y: 0; --tw-translate-x: 0; --tw-translate-y: 0; background-color: white; color: #404040; font-family: "SF Pro Display", -apple-system, BlinkMacSystemFont, Inter, "Segoe UI", Roboto, Helvetica, Arial, sans-serif, "Apple Color Emoji", "Segoe UI Emoji", "Segoe UI Symbol";">enerative </span><strong style="--tw-border-spacing-x: 0; --tw-border-spacing-y: 0; --tw-ring-color: rgb(59 130 246 / 0.5); --tw-ring-offset-color: #fff; --tw-ring-offset-shadow: 0 0 #0000; --tw-ring-offset-width: 0px; --tw-ring-shadow: 0 0 #0000; --tw-rotate: 0; --tw-scale-x: 1; --tw-scale-y: 1; --tw-scroll-snap-strictness: proximity; --tw-shadow-colored: 0 0 #0000; --tw-shadow: 0 0 #0000; --tw-skew-x: 0; --tw-skew-y: 0; --tw-translate-x: 0; --tw-translate-y: 0; background-color: white; color: #404040; font-family: "SF Pro Display", -apple-system, BlinkMacSystemFont, Inter, "Segoe UI", Roboto, Helvetica, Arial, sans-serif, "Apple Color Emoji", "Segoe UI Emoji", "Segoe UI Symbol";">P</strong><span style="--tw-border-spacing-x: 0; --tw-border-spacing-y: 0; --tw-ring-color: rgb(59 130 246 / 0.5); --tw-ring-offset-color: #fff; --tw-ring-offset-shadow: 0 0 #0000; --tw-ring-offset-width: 0px; --tw-ring-shadow: 0 0 #0000; --tw-rotate: 0; --tw-scale-x: 1; --tw-scale-y: 1; --tw-scroll-snap-strictness: proximity; --tw-shadow-colored: 0 0 #0000; --tw-shadow: 0 0 #0000; --tw-skew-x: 0; --tw-skew-y: 0; --tw-translate-x: 0; --tw-translate-y: 0; background-color: white; color: #404040; font-family: "SF Pro Display", -apple-system, BlinkMacSystemFont, Inter, "Segoe UI", Roboto, Helvetica, Arial, sans-serif, "Apple Color Emoji", "Segoe UI Emoji", "Segoe UI Symbol";">re-trained </span><strong style="--tw-border-spacing-x: 0; --tw-border-spacing-y: 0; --tw-ring-color: rgb(59 130 246 / 0.5); --tw-ring-offset-color: #fff; --tw-ring-offset-shadow: 0 0 #0000; --tw-ring-offset-width: 0px; --tw-ring-shadow: 0 0 #0000; --tw-rotate: 0; --tw-scale-x: 1; --tw-scale-y: 1; --tw-scroll-snap-strictness: proximity; --tw-shadow-colored: 0 0 #0000; --tw-shadow: 0 0 #0000; --tw-skew-x: 0; --tw-skew-y: 0; --tw-translate-x: 0; --tw-translate-y: 0; background-color: white; color: #404040; font-family: "SF Pro Display", -apple-system, BlinkMacSystemFont, Inter, "Segoe UI", Roboto, Helvetica, Arial, sans-serif, "Apple Color Emoji", "Segoe UI Emoji", "Segoe UI Symbol";">T</strong></u><span style="--tw-border-spacing-x: 0; --tw-border-spacing-y: 0; --tw-ring-color: rgb(59 130 246 / 0.5); --tw-ring-offset-color: #fff; --tw-ring-offset-shadow: 0 0 #0000; --tw-ring-offset-width: 0px; --tw-ring-shadow: 0 0 #0000; --tw-rotate: 0; --tw-scale-x: 1; --tw-scale-y: 1; --tw-scroll-snap-strictness: proximity; --tw-shadow-colored: 0 0 #0000; --tw-shadow: 0 0 #0000; --tw-skew-x: 0; --tw-skew-y: 0; --tw-translate-x: 0; --tw-translate-y: 0; background-color: white; color: #404040; font-family: "SF Pro Display", -apple-system, BlinkMacSystemFont, Inter, "Segoe UI", Roboto, Helvetica, Arial, sans-serif, "Apple Color Emoji", "Segoe UI Emoji", "Segoe UI Symbol";"><u>ransformer)</u> series are decoder-only models pretrained on large-scale unsupervised text data and finetuned for specific tasks such as text classification, sentiment analysis, question-answering, and summarization. The GPT models, including GPT-2, (</span><em style="--tw-border-spacing-x: 0; --tw-border-spacing-y: 0; --tw-ring-color: rgb(59 130 246 / 0.5); --tw-ring-offset-color: #fff; --tw-ring-offset-shadow: 0 0 #0000; --tw-ring-offset-width: 0px; --tw-ring-shadow: 0 0 #0000; --tw-rotate: 0; --tw-scale-x: 1; --tw-scale-y: 1; --tw-scroll-snap-strictness: proximity; --tw-shadow-colored: 0 0 #0000; --tw-shadow: 0 0 #0000; --tw-skew-x: 0; --tw-skew-y: 0; --tw-translate-x: 0; --tw-translate-y: 0; background-color: white; color: #404040; font-family: "SF Pro Display", -apple-system, BlinkMacSystemFont, Inter, "Segoe UI", Roboto, Helvetica, Arial, sans-serif, "Apple Color Emoji", "Segoe UI Emoji", "Segoe UI Symbol";"><a href="https://arxiv.org/abs/2005.14165" rel="" style="--tw-border-spacing-x: 0; --tw-border-spacing-y: 0; --tw-ring-color: rgb(59 130 246 / 0.5); --tw-ring-offset-color: #fff; --tw-ring-offset-shadow: 0 0 #0000; --tw-ring-offset-width: 0px; --tw-ring-shadow: 0 0 #0000; --tw-rotate: 0; --tw-scale-x: 1; --tw-scale-y: 1; --tw-scroll-snap-strictness: proximity; --tw-shadow-colored: 0 0 #0000; --tw-shadow: 0 0 #0000; --tw-skew-x: 0; --tw-skew-y: 0; --tw-translate-x: 0; --tw-translate-y: 0;">GPT-3 Language Models are Few-Shot Learners, 2020</a></em><span style="--tw-border-spacing-x: 0; --tw-border-spacing-y: 0; --tw-ring-color: rgb(59 130 246 / 0.5); --tw-ring-offset-color: #fff; --tw-ring-offset-shadow: 0 0 #0000; --tw-ring-offset-width: 0px; --tw-ring-shadow: 0 0 #0000; --tw-rotate: 0; --tw-scale-x: 1; --tw-scale-y: 1; --tw-scroll-snap-strictness: proximity; --tw-shadow-colored: 0 0 #0000; --tw-shadow: 0 0 #0000; --tw-skew-x: 0; --tw-skew-y: 0; --tw-translate-x: 0; --tw-translate-y: 0; background-color: white; color: #404040; font-family: "SF Pro Display", -apple-system, BlinkMacSystemFont, Inter, "Segoe UI", Roboto, Helvetica, Arial, sans-serif, "Apple Color Emoji", "Segoe UI Emoji", "Segoe UI Symbol";">), and the more recent GPT-4, have shown remarkable performance in various benchmarks and are currently the most popular architecture for natural language processing.</span></div><div><br /></div></div><div><u style="font-weight: bold;"><br /></u></div><div><span style="font-weight: bold;"> <u>1.3.3 Encoder-Decoder :</u></span><span style="font-weight: bold;"> Compacts input into an output. </span><span style="font-weight: 700;">Ex : French to English translation </span></div><div> - The original paper creator transformer architecture</div><div> - Translation tasks, cross attention</div><div><br /></div><div><p style="--tw-border-spacing-x: 0; --tw-border-spacing-y: 0; --tw-ring-color: rgb(59 130 246 / 0.5); --tw-ring-offset-color: #fff; --tw-ring-offset-shadow: 0 0 #0000; --tw-ring-offset-width: 0px; --tw-ring-shadow: 0 0 #0000; --tw-rotate: 0; --tw-scale-x: 1; --tw-scale-y: 1; --tw-scroll-snap-strictness: proximity; --tw-shadow-colored: 0 0 #0000; --tw-shadow: 0 0 #0000; --tw-skew-x: 0; --tw-skew-y: 0; --tw-translate-x: 0; --tw-translate-y: 0; background-color: white; color: #404040; font-family: "SF Pro Display", -apple-system, BlinkMacSystemFont, Inter, "Segoe UI", Roboto, Helvetica, Arial, sans-serif, "Apple Color Emoji", "Segoe UI Emoji", "Segoe UI Symbol"; line-height: 1.6em; margin: 0 0 var(--size-20) 0;">Encoder-decoder models are typically used for natural language processing tasks that involve understanding input sequences and generating output sequences, often with different lengths and structures. They are particularly good at tasks where there is a complex mapping between the input and output sequences and where it is crucial to capture the relationships between the elements in both sequences. Some common use cases for encoder-decoder models include text translation and summarization.</p><p style="--tw-border-spacing-x: 0; --tw-border-spacing-y: 0; --tw-ring-color: rgb(59 130 246 / 0.5); --tw-ring-offset-color: #fff; --tw-ring-offset-shadow: 0 0 #0000; --tw-ring-offset-width: 0px; --tw-ring-shadow: 0 0 #0000; --tw-rotate: 0; --tw-scale-x: 1; --tw-scale-y: 1; --tw-scroll-snap-strictness: proximity; --tw-shadow-colored: 0 0 #0000; --tw-shadow: 0 0 #0000; --tw-skew-x: 0; --tw-skew-y: 0; --tw-translate-x: 0; --tw-translate-y: 0; background-color: white; color: #404040; font-family: "SF Pro Display", -apple-system, BlinkMacSystemFont, Inter, "Segoe UI", Roboto, Helvetica, Arial, sans-serif, "Apple Color Emoji", "Segoe UI Emoji", "Segoe UI Symbol"; line-height: 1.6em; margin: 0 0 var(--size-20) 0;"><br /></p><p style="--tw-border-spacing-x: 0; --tw-border-spacing-y: 0; --tw-ring-color: rgb(59 130 246 / 0.5); --tw-ring-offset-color: #fff; --tw-ring-offset-shadow: 0 0 #0000; --tw-ring-offset-width: 0px; --tw-ring-shadow: 0 0 #0000; --tw-rotate: 0; --tw-scale-x: 1; --tw-scale-y: 1; --tw-scroll-snap-strictness: proximity; --tw-shadow-colored: 0 0 #0000; --tw-shadow: 0 0 #0000; --tw-skew-x: 0; --tw-skew-y: 0; --tw-translate-x: 0; --tw-translate-y: 0; background-color: white; color: #404040; font-family: "SF Pro Display", -apple-system, BlinkMacSystemFont, Inter, "Segoe UI", Roboto, Helvetica, Arial, sans-serif, "Apple Color Emoji", "Segoe UI Emoji", "Segoe UI Symbol"; line-height: 1.6em; margin: 0 0 var(--size-20) 0;">Some notable examples of these new encoder-decoder models include</p><ul style="--tw-border-spacing-x: 0; --tw-border-spacing-y: 0; --tw-ring-color: rgb(59 130 246 / 0.5); --tw-ring-offset-color: #fff; --tw-ring-offset-shadow: 0 0 #0000; --tw-ring-offset-width: 0px; --tw-ring-shadow: 0 0 #0000; --tw-rotate: 0; --tw-scale-x: 1; --tw-scale-y: 1; --tw-scroll-snap-strictness: proximity; --tw-shadow-colored: 0 0 #0000; --tw-shadow: 0 0 #0000; --tw-skew-x: 0; --tw-skew-y: 0; --tw-translate-x: 0; --tw-translate-y: 0; background-color: white; color: #404040; font-family: "SF Pro Display", -apple-system, BlinkMacSystemFont, Inter, "Segoe UI", Roboto, Helvetica, Arial, sans-serif, "Apple Color Emoji", "Segoe UI Emoji", "Segoe UI Symbol"; margin-top: 0px; padding: 0px;"><li style="--tw-border-spacing-x: 0; --tw-border-spacing-y: 0; --tw-ring-color: rgb(59 130 246 / 0.5); --tw-ring-offset-color: #fff; --tw-ring-offset-shadow: 0 0 #0000; --tw-ring-offset-width: 0px; --tw-ring-shadow: 0 0 #0000; --tw-rotate: 0; --tw-scale-x: 1; --tw-scale-y: 1; --tw-scroll-snap-strictness: proximity; --tw-shadow-colored: 0 0 #0000; --tw-shadow: 0 0 #0000; --tw-skew-x: 0; --tw-skew-y: 0; --tw-translate-x: 0; --tw-translate-y: 0; margin: var(--size-8) 0 0 var(--size-32);"><p style="--tw-border-spacing-x: 0; --tw-border-spacing-y: 0; --tw-ring-color: rgb(59 130 246 / 0.5); --tw-ring-offset-color: #fff; --tw-ring-offset-shadow: 0 0 #0000; --tw-ring-offset-width: 0px; --tw-ring-shadow: 0 0 #0000; --tw-rotate: 0; --tw-scale-x: 1; --tw-scale-y: 1; --tw-scroll-snap-strictness: proximity; --tw-shadow-colored: 0 0 #0000; --tw-shadow: 0 0 #0000; --tw-skew-x: 0; --tw-skew-y: 0; --tw-translate-x: 0; --tw-translate-y: 0; box-sizing: border-box; color: var(--print_on_web_bg_color, var(--color-primary)); line-height: 1.6em; margin: 0px; padding-left: var(--size-4);"><span style="--tw-border-spacing-x: 0; --tw-border-spacing-y: 0; --tw-ring-color: rgb(59 130 246 / 0.5); --tw-ring-offset-color: #fff; --tw-ring-offset-shadow: 0 0 #0000; --tw-ring-offset-width: 0px; --tw-ring-shadow: 0 0 #0000; --tw-rotate: 0; --tw-scale-x: 1; --tw-scale-y: 1; --tw-scroll-snap-strictness: proximity; --tw-shadow-colored: 0 0 #0000; --tw-shadow: 0 0 #0000; --tw-skew-x: 0; --tw-skew-y: 0; --tw-translate-x: 0; --tw-translate-y: 0;">BART (</span><em style="--tw-border-spacing-x: 0; --tw-border-spacing-y: 0; --tw-ring-color: rgb(59 130 246 / 0.5); --tw-ring-offset-color: #fff; --tw-ring-offset-shadow: 0 0 #0000; --tw-ring-offset-width: 0px; --tw-ring-shadow: 0 0 #0000; --tw-rotate: 0; --tw-scale-x: 1; --tw-scale-y: 1; --tw-scroll-snap-strictness: proximity; --tw-shadow-colored: 0 0 #0000; --tw-shadow: 0 0 #0000; --tw-skew-x: 0; --tw-skew-y: 0; --tw-translate-x: 0; --tw-translate-y: 0;"><a href="https://arxiv.org/abs/1910.13461" rel="" style="--tw-border-spacing-x: 0; --tw-border-spacing-y: 0; --tw-ring-color: rgb(59 130 246 / 0.5); --tw-ring-offset-color: #fff; --tw-ring-offset-shadow: 0 0 #0000; --tw-ring-offset-width: 0px; --tw-ring-shadow: 0 0 #0000; --tw-rotate: 0; --tw-scale-x: 1; --tw-scale-y: 1; --tw-scroll-snap-strictness: proximity; --tw-shadow-colored: 0 0 #0000; --tw-shadow: 0 0 #0000; --tw-skew-x: 0; --tw-skew-y: 0; --tw-translate-x: 0; --tw-translate-y: 0;">Denoising Sequence-to-Sequence Pre-training for Natural Language Generation, Translation, and Comprehension, 2019</a></em><span style="--tw-border-spacing-x: 0; --tw-border-spacing-y: 0; --tw-ring-color: rgb(59 130 246 / 0.5); --tw-ring-offset-color: #fff; --tw-ring-offset-shadow: 0 0 #0000; --tw-ring-offset-width: 0px; --tw-ring-shadow: 0 0 #0000; --tw-rotate: 0; --tw-scale-x: 1; --tw-scale-y: 1; --tw-scroll-snap-strictness: proximity; --tw-shadow-colored: 0 0 #0000; --tw-shadow: 0 0 #0000; --tw-skew-x: 0; --tw-skew-y: 0; --tw-translate-x: 0; --tw-translate-y: 0;">) </span><span style="--tw-border-spacing-x: 0; --tw-border-spacing-y: 0; --tw-ring-color: rgb(59 130 246 / 0.5); --tw-ring-offset-color: #fff; --tw-ring-offset-shadow: 0 0 #0000; --tw-ring-offset-width: 0px; --tw-ring-shadow: 0 0 #0000; --tw-rotate: 0; --tw-scale-x: 1; --tw-scale-y: 1; --tw-scroll-snap-strictness: proximity; --tw-shadow-colored: 0 0 #0000; --tw-shadow: 0 0 #0000; --tw-skew-x: 0; --tw-skew-y: 0; --tw-translate-x: 0; --tw-translate-y: 0; color: var(--print_on_web_bg_color, var(--color-primary));">and T5 (</span><em style="--tw-border-spacing-x: 0; --tw-border-spacing-y: 0; --tw-ring-color: rgb(59 130 246 / 0.5); --tw-ring-offset-color: #fff; --tw-ring-offset-shadow: 0 0 #0000; --tw-ring-offset-width: 0px; --tw-ring-shadow: 0 0 #0000; --tw-rotate: 0; --tw-scale-x: 1; --tw-scale-y: 1; --tw-scroll-snap-strictness: proximity; --tw-shadow-colored: 0 0 #0000; --tw-shadow: 0 0 #0000; --tw-skew-x: 0; --tw-skew-y: 0; --tw-translate-x: 0; --tw-translate-y: 0; color: var(--print_on_web_bg_color, var(--color-primary));"><a href="https://arxiv.org/abs/1910.10683" rel="" style="--tw-border-spacing-x: 0; --tw-border-spacing-y: 0; --tw-ring-color: rgb(59 130 246 / 0.5); --tw-ring-offset-color: #fff; --tw-ring-offset-shadow: 0 0 #0000; --tw-ring-offset-width: 0px; --tw-ring-shadow: 0 0 #0000; --tw-rotate: 0; --tw-scale-x: 1; --tw-scale-y: 1; --tw-scroll-snap-strictness: proximity; --tw-shadow-colored: 0 0 #0000; --tw-shadow: 0 0 #0000; --tw-skew-x: 0; --tw-skew-y: 0; --tw-translate-x: 0; --tw-translate-y: 0;">Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer, 2019</a></em><span style="--tw-border-spacing-x: 0; --tw-border-spacing-y: 0; --tw-ring-color: rgb(59 130 246 / 0.5); --tw-ring-offset-color: #fff; --tw-ring-offset-shadow: 0 0 #0000; --tw-ring-offset-width: 0px; --tw-ring-shadow: 0 0 #0000; --tw-rotate: 0; --tw-scale-x: 1; --tw-scale-y: 1; --tw-scroll-snap-strictness: proximity; --tw-shadow-colored: 0 0 #0000; --tw-shadow: 0 0 #0000; --tw-skew-x: 0; --tw-skew-y: 0; --tw-translate-x: 0; --tw-translate-y: 0; color: var(--print_on_web_bg_color, var(--color-primary));">).</span></p></li></ul></div><div><br /></div><div>*Ref: </div><div>https://magazine.sebastianraschka.com/p/understanding-encoder-and-decoder</div><div><br /></div><div><br /></div><div><b><u>1.5 LLMs are based on the three types of building blocks</u></b> </div><div><br /></div><div><span style="font-weight: bold;"> <u>1.5.1 Attention: </u></span> </div><div><br /></div><div><div class="content" tabindex="0"><div class="ac-container ac-adaptiveCard" streaming=""><div class="ac-textBlock"><p>In the context of LLM, <strong>attention</strong> is defined as a mechanism that allows the model to selectively focus on different parts of the input text. This mechanism helps the model attend to the input text’s most relevant parts and generate more accurate predictions</p><p>The use of attention in LLMs is to improve the model’s ability to understand the context of the input text and generate more coherent and relevant output. Attention mechanisms in LLMs, particularly the self-attention mechanism used in transformers, allow the model to weigh the importance of different words or phrases in a given context.</p><p>There are two types of attention mechanisms in LLMs: <strong>self-attention</strong> and <strong>cross-attention</strong>. </p><p><b><span> </span>Self-attention</b> is used to weigh the importance of different words or phrases within the same input text, </p><p><b><span> </span>Cross-attention</b> is used to weigh the importance of different words or phrases between two different input texts.</p><p>The measurement of attention in LLMs is done by calculating the attention weights assigned to each word or phrase in the input text. These weights are calculated using a softmax function, which normalizes the weights and ensures that they sum up to 1</p><p>Here are a couple of examples of how attention is used in LLMs:</p><ol><li><a class="tooltip-target" data-citationid="17d67315-b059-f527-3148-7c4069fcdf81-26-group" h="ID=SERP,5026.1" href="https://www.analyticsvidhya.com/blog/2023/03/an-introduction-to-large-language-models-llms/" target="_blank">In <strong>machine translation</strong>, attention is used to align the source and target sentences and generate more accurate translations </a></li><li><a class="tooltip-target" data-citationid="17d67315-b059-f527-3148-7c4069fcdf81-30-group" h="ID=SERP,5026.1" href="https://www.analyticsvidhya.com/blog/2023/03/an-introduction-to-large-language-models-llms/">In <strong>question answering</strong>, attention is used to identify the most relevant parts of the input text that can help answer the question </a></li></ol></div><div aria-hidden="true" class="ac-horizontal-separator"></div></div><cib-overlay></cib-overlay></div></div><div><br /></div><div><br /></div><div><span style="font-weight: bold;"> <u>1.5.2 Parallelism and </u></span><u style="font-weight: bold;">Scalability: </u></div><div><br /></div><div><div class="content" tabindex="0"><div class="ac-container ac-adaptiveCard"><div class="ac-textBlock"><p><strong>LLM</strong> stands for <strong>Large Language Model</strong>. It is a machine learning model that is trained on large amounts of data to generate text. LLMs are used in various natural language processing tasks such as language translation, text summarization, and question answering .</p><p><strong>Parallelism</strong> is used to train the model faster by distributing the workload across multiple processors or GPUs. There are two types of parallelism: <strong>data parallelism</strong> and <strong>model parallelism</strong>.</p><ul><li><p><strong>Data parallelism</strong> involves splitting the data into smaller batches and processing them in parallel across multiple processors or GPUs. This technique is useful when the model is too large to fit into a single GPU memory.</p></li><li><p><strong>Model parallelism</strong> involves splitting the model into smaller parts and processing them in parallel across multiple processors or GPUs. This technique is useful when the model is too large to fit into a single processor or GPU memory.</p></li></ul><p><strong>Scalability</strong> is used to train the model on larger datasets or with more complex architectures. Scalability can be measured in terms of <strong>speedup</strong> and <strong>efficiency</strong>.</p><ul><li><p><strong>Speedup</strong> is the ratio of the time taken to complete a task on a single processor or GPU to the time taken to complete the same task on multiple processors or GPUs. A higher speedup indicates better scalability.</p></li><li><p><strong>Efficiency</strong> is the ratio of the speedup to the number of processors or GPUs used. A higher efficiency indicates better scalability.</p></li></ul><p>Here are a couple of examples of LLMs:</p><ol><li><p><strong>GPT-3</strong>: It is a state-of-the-art LLM developed by OpenAI that has 175 billion parameters. <a class="tooltip-target" data-citationid="fb7a06c5-37f9-9185-ebd4-a81957886899-45-group" h="ID=SERP,5026.1" href="https://www.computerworld.com/article/3697649/what-are-large-language-models-and-how-are-they-used-in-generative-ai.html">It is used for various natural language processing tasks such as language translation, text summarization, and question answering </a><a aria-label="1: GPT-3" class="ac-anchor sup-target" data-citationid="fb7a06c5-37f9-9185-ebd4-a81957886899-45" h="ID=SERP,5026.1" href="https://www.computerworld.com/article/3697649/what-are-large-language-models-and-how-are-they-used-in-generative-ai.html" target="_blank"><sup>1</sup></a>.</p></li><li><p><strong>BERT</strong>: It is another popular LLM developed by Google that has 340 million parameters. <a class="tooltip-target" data-citationid="fb7a06c5-37f9-9185-ebd4-a81957886899-50-group" h="ID=SERP,5026.1" href="https://www.computerworld.com/article/3697649/what-are-large-language-models-and-how-are-they-used-in-generative-ai.html">It is used for various natural language processing tasks such as sentiment analysis, named entity recognition, and question answering </a><a aria-label="1: BERT" class="ac-anchor sup-target" data-citationid="fb7a06c5-37f9-9185-ebd4-a81957886899-50" h="ID=SERP,5026.1" href="https://www.computerworld.com/article/3697649/what-are-large-language-models-and-how-are-they-used-in-generative-ai.html" target="_blank"><sup>1</sup></a>.</p></li></ol></div><div aria-hidden="true" class="ac-horizontal-separator"></div></div><cib-overlay></cib-overlay></div></div><div><br /></div><div> <span style="font-weight: bold;"> <u>1.5.3 Sequence Modeling: </u></span> </div><div><u style="font-weight: bold;"><br /></u></div><div><br /></div><div><b><u>2. GenAI - Popular ones being used in 2023-24 </u></b>: </div><p></p><p><br /></p><p></p><div class="separator" style="clear: both; text-align: center;"><a href="https://blogger.googleusercontent.com/img/a/AVvXsEipXOPtSfTubXuhKKRQWQvuEvBQs-1e0XNPK_uBV18CfLLK1L0VUVcz6u1UOZFR-fv2LLv0d9o7NLfT0imrCDBawX8mC0ZvyKGihGObcFWZ1JtHG7LwuOx1HRUbD1oTk_VWlWTGz8oa9odkxIhFVgIS-VFTu45bqk9iyKsoqjUYHeJasfVeCqDfnisoq3Q" style="clear: left; float: left; margin-bottom: 1em; margin-left: 1em;"><img alt="" data-original-height="235" data-original-width="981" height="142" src="https://blogger.googleusercontent.com/img/a/AVvXsEipXOPtSfTubXuhKKRQWQvuEvBQs-1e0XNPK_uBV18CfLLK1L0VUVcz6u1UOZFR-fv2LLv0d9o7NLfT0imrCDBawX8mC0ZvyKGihGObcFWZ1JtHG7LwuOx1HRUbD1oTk_VWlWTGz8oa9odkxIhFVgIS-VFTu45bqk9iyKsoqjUYHeJasfVeCqDfnisoq3Q=w590-h142" width="590" /></a></div><br /><br /><p></p><p></p><div class="separator" style="clear: both; text-align: left;">You can play around the GenAI Use cases in one place with example, you can try <a href="http://playground.katonic.ai">playground.katonic.ai</a></div><br />In tale above, what we are saying is LLaMA is a good open source model for text output. codellama-7b-instruct has 7billion constructs that can be used for writing code. <div><br /></div><div><b><u>Prompts</u></b> are inputs that can be asked to LLM for desired output. You have to use inherent prompt inputs for best output. This is an engineering discipline called <u>Prompt Engineering</u>.</div><div><br /></div><div><b><u>3. What does Generative AI do?</u></b></div><div><b><u><br /></u></b></div><div>Traditional AI used to take time in development, iterations, deployment, consumption, data training etc.</div><div>Gen AI can be quickly standing in a matter of weeks. </div><div><br /></div><div>Gen AI can basically do one of the following functions and some examples:-</div><div><br /></div><div><b>3.1. Summarization </b>: Regulatory Guidelines/ Risk reports/ UW/Claims/policy/ Corporate Functions</div><div><b>3.2. Reference & Co-Pilot </b>: Extract key information like Information extraction/ Risk Analysis/ Sentiment mining/ Fraud/ Event detection/ Web mining/ CX/CJ Insights</div><div><b>3.3. Expansion: </b>Automated mails/ descriptions/ qualitative reports/ Synthetic data/ Advisory-B2B/ B2C</div><div><b>3.4. Transformation </b>: Change Language structure Ex Translation/ Code writing/ Data format change/ Tone change/ AI driven BI</div><div><br /></div><div><b>4. Ways to use LLM APIs</b></div><div><ul style="text-align: left;"><li>Integrate to other apps</li><li>Virtual Assistants</li><li>Developer Co-Pilot</li><li>Custom Applications</li></ul><div><b>5. Example Use case from Insurance:-</b></div></div><div><ul style="text-align: left;"><li>Content Generation: Lead generation/ Onboarding/ Customer Management/ Delinquency & Foreclosures</li><li>Workflow management: </li><li>Client Experience and Interaction</li><li>Security Compliance</li><li>Workflow Optimization</li></ul><div><b>6. There are 2 broad patterns in which use cases fall:-</b></div></div><div><b><br /></b></div><div><b>6.1. Retrieval Augmented Generation (RAG)- </b>Retrieve and answer in context (<b>in-context learning)</b></div><div><b>Agent based architecture/ Fine tuning for optimization/ </b></div><div><br /></div><div>Question/Task > LLM > (Indexing) Indexed Query > Vector Store > <b>Give Context</b> > Contextual Prompt Creation > LLM> Output > Output parsing> Answer </div><div><br /></div><div>LLM is fixed/OOB domain agnostic GPT3/4. You can give context to ask the question.</div><div>Context can be given in <b>zero shot learning</b>/ <b>few shot learning</b> (like examples of desired output).</div><div>No data leakage problem. </div><div><br /></div><div>If you pass data directly (fine-tuning -> given this context, I need examples of right answer, then model aligns) into LLM, it can give direct answers. but it has a security challenge. Fine tuning increases accuracy and alignment. But data can become stale, as it's not always live. It is recommended to go with RAG which is outside the LLM model, instead of fine tuning the model through your own data.</div><div><b><br /></b></div><div><b>So in the Underwriting example:-</b></div><div> <b>a) Agent-Assistance :</b> smart bot to enable agent answer diff customer/ prospect questions, Tap into existing contractual docs, guidelines, benefits, calculations, research & insights ex Knowledge management, chatbot,...</div><div> <b>b) Underwriting/Risk Co-pilots for Mortgage : </b>The co-pilot helps underwriter go through several steps to assess risk, summarize the qualifying income through reviewing and identifying various sources of income, do some appraisals, contextualize credit and assets. LLM can curate much better than analytic AI. Automatically analyze live data, write output, and generally help the underwriter </div><div><br /></div><div><b>6.2. Multi-Hop/ Multi-stage Problem Solving :</b> </div><div><div> <b>Insight Agents:</b> conversational business intelligence, data quality assurance, analytics and insights co-pilot, decision support agent.</div></div><div><br /></div><div>Multi-Hop problem solving is different from RAG. Some large LLMs have logic and reasoning built. You can build agents to build say conversational BI instead of conversational Bot, which is fully textual. Multi-hop can build insights, cross match data, chain of thoughts etc. Reason and Act - REACT is a sequence of steps to accomplish a task before output can be given. More complex reasoning can also be tried. You can need graphical analysis example concentration of claims that can point to one garage , or relationship between claims... Knowledge graphs between data sets can throw up unique intelligence based on simple textual question prompts. Can help in decision support systems. </div><div><br /></div><div><br /></div><div><b>7. Some concepts to know here:-</b></div><div><b>- May require fine tuning or reform of the RAG : </b>To change the model as per your data. Here, the use your own data sets to train the model. Private Trainership is available from GPT4.<b> </b></div><div><b><span><br /></span></b></div><div><b><span> </span><span> </span>-<span> </span>SFT and RLHF</b></div><div><div class="content" tabindex="0"><div class="ac-container ac-adaptiveCard" streaming=""><div class="ac-textBlock"><p>In the context of LLMs, <strong>SFT</strong> stands for <strong>Supervised Fine-Tuning</strong>. It is a technique used to fine-tune a pre-trained LLM on a specific task by providing it with labeled examples.</p><p><strong>RLHF</strong> stands for <strong>Reinforcement Learning from Human Feedback</strong>. It is a method used to train LLMs to align their output with human intentions and values. RLHF involves teaching an LLM to understand human preferences by assigning scores to different responses from the base model. The goal is to use the preference model to alter the behavior of the base model in response to a prompt.</p><p>Here are some examples of how RLHF is used in LLMs:</p><ol><li><p><strong>ChatGPT</strong>: It is a state-of-the-art LLM developed by AssemblyAI that uses RLHF to learn human preferences and provide a more controlled user experience.</p></li><li><p><strong>NVIDIA SteerLM</strong>: It is a technique developed by NVIDIA that uses RLHF to customize LLMs during inference.</p></li></ol></div><div aria-hidden="true" class="ac-horizontal-separator"></div></div><cib-overlay></cib-overlay></div></div><div><div><b><br /></b></div><div><b>- Foundational Model shift </b></div><div><b><br /></b></div><div><b>- If trainings are not proper, LLM can hallucinate, missing context.</b></div></div><div><b><br /></b></div><div><b>8. SUMMARY of ARCHITECTURES</b></div><div><b>For completeness of text, let me mention that these are the main Attention-based Architectures:-</b></div><div><b>8.1. Encoder-Decoder (RNN based)</b></div><div><b>8.2. Transformer Model (non-RNN)</b></div><div><b>8.3. Graph Neural Networks (GNN)</b></div><div><b>8.4. Memory Augmented Neural Networks</b></div><div><b><br /></b></div><div><p style="background: 0px 0px rgb(255, 255, 255); border: 0px; color: #555555; font-family: "Helvetica Neue", Helvetica, sans-serif; font-feature-settings: normal; font-kerning: auto; font-optical-sizing: auto; font-size: 15px; font-stretch: normal; font-style: italic; font-variant-alternates: normal; font-variant-east-asian: normal; font-variant-numeric: normal; font-variant-position: normal; font-variation-settings: normal; line-height: 1.5em; margin: 0px 0px 1.2em; outline: 0px; padding: 0px; vertical-align: baseline;"><i style="background: 0px 0px; border: 0px; margin: 0px; outline: 0px; padding: 0px; vertical-align: baseline;">A graph is a versatile data structure that lends itself well to the way data is organized in many real-world scenarios. </i><i style="background-attachment: initial; background-clip: initial; background-image: initial; background-origin: initial; background-position: 0px 0px; background-repeat: initial; background-size: initial; border: 0px; margin: 0px; outline: 0px; padding: 0px; vertical-align: baseline;">We can think of an image as a graph, where each pixel is a node, directly connected to its neighboring pixels …</i></p><p style="background: 0px 0px rgb(255, 255, 255); border: 0px; color: #555555; font-family: "Helvetica Neue", Helvetica, sans-serif; font-feature-settings: normal; font-kerning: auto; font-optical-sizing: auto; font-size: 15px; font-stretch: normal; font-style: italic; font-variant-alternates: normal; font-variant-east-asian: normal; font-variant-numeric: normal; font-variant-position: normal; font-variation-settings: normal; line-height: 1.5em; margin: 0px 0px 1.2em; outline: 0px; padding: 0px; text-align: right; vertical-align: baseline;">– <a href="https://www.amazon.com/Advanced-Deep-Learning-Python-next-generation/dp/178995617X" style="background: 0px 0px; border: 0px; color: #428bca; margin: 0px; outline: 0px; padding: 0px; text-decoration-line: none; vertical-align: baseline;">Advanced Deep Learning with Python</a>, 2019.</p></div><div><span style="background-color: white; color: #555555; font-family: "Helvetica Neue", Helvetica, sans-serif; font-size: 15px;">Of particular interest are the </span><i style="background: 0px 0px rgb(255, 255, 255); border: 0px; color: #555555; font-family: "Helvetica Neue", Helvetica, sans-serif; font-size: 15px; margin: 0px; outline: 0px; padding: 0px; vertical-align: baseline;"><b>Graph Attention Networks</b></i><span style="background-color: white; color: #555555; font-family: "Helvetica Neue", Helvetica, sans-serif; font-size: 15px;"><b> (GAT)</b> that employ a self-attention mechanism within a <b>graph convolutional network (GCN)</b>, where the latter updates the state vectors by performing a convolution over the nodes of the graph.</span></div><div><b><br /></b></div><div><p style="background: 0px 0px rgb(255, 255, 255); border: 0px; color: #555555; font-family: "Helvetica Neue", Helvetica, sans-serif; font-feature-settings: normal; font-kerning: auto; font-optical-sizing: auto; font-size: 15px; font-stretch: normal; font-variant-alternates: normal; font-variant-east-asian: normal; font-variant-numeric: normal; font-variant-position: normal; font-variation-settings: normal; line-height: 1.5em; margin: 0px 0px 1.2em; outline: 0px; padding: 0px; vertical-align: baseline;">In the encoder-decoder attention-based architectures, the set of vectors that encode the input sequence can be considered external memory, to which the encoder writes and from which the decoder reads. However, a limitation arises because the encoder can only write to this memory, and the decoder can only read.</p><p style="background: 0px 0px rgb(255, 255, 255); border: 0px; color: #555555; font-family: "Helvetica Neue", Helvetica, sans-serif; font-feature-settings: normal; font-kerning: auto; font-optical-sizing: auto; font-size: 15px; font-stretch: normal; font-variant-alternates: normal; font-variant-east-asian: normal; font-variant-numeric: normal; font-variant-position: normal; font-variation-settings: normal; line-height: 1.5em; margin: 0px 0px 1.2em; outline: 0px; padding: 0px; vertical-align: baseline;"><b>Memory-Augmented Neural Networks (MANNs)</b> are recent algorithms that aim to address this limitation.<span class="Apple-converted-space" style="background: 0px 0px; border: 0px; margin: 0px; outline: 0px; padding: 0px; vertical-align: baseline;"> </span></p><p style="background: 0px 0px rgb(255, 255, 255); border: 0px; color: #555555; font-family: "Helvetica Neue", Helvetica, sans-serif; font-feature-settings: normal; font-kerning: auto; font-optical-sizing: auto; font-size: 15px; font-stretch: normal; font-variant-alternates: normal; font-variant-east-asian: normal; font-variant-numeric: normal; font-variant-position: normal; font-variation-settings: normal; line-height: 1.5em; margin: 0px 0px 1.2em; outline: 0px; padding: 0px; vertical-align: baseline;">The <b>Neural Turing Machine (NTM)</b> is one type of MANN. It consists of a neural network controller that takes an input to produce an output and performs read and write operations to memory. Examples of applications for MANNs include question-answering and chat bots, where an external memory stores a large database of sequences (or facts) that the neural network taps into. </p></div><div><b><br /></b></div><div><b>9. Policy and Principles around GenAI:-</b></div><div><span> </span>Guardrails, policy enforcements, or Governance regarding inputs to GenAI is an open subject. There could be LLM tools or technology that can enforce these policy guardrails that can ensure security,privacy etc.</div>Unknownnoreply@blogger.com0tag:blogger.com,1999:blog-6698881918084302179.post-16493721481758656082023-10-29T14:31:00.001-07:002023-10-29T14:31:18.369-07:00The Impact of Artificial Intelligence on Teaching, Learning, and Educational Equity: A Review<div>Title: The Impact of Artificial Intelligence on Teaching, Learning, and Educational Equity: A Review<br></div><div><br></div><div><b>Abstract</b></div><div><br></div><div>This paper provides a comprehensive review of the impact of artificial intelligence (AI) on education, with a focus on teaching, learning, and educational equity. AI refers to computer systems that exhibit human-like intelligence and capabilities. The rapid advancement of AI, fueled by increases in computing power and availability of big data, has enabled the proliferation of AI applications in education. However, as AI becomes more deeply embedded in educational technologies, significant opportunities as well as risks emerge. </div><div><br></div><div>This paper reviews recent literature on AI in education and synthesizes key insights around three major themes: </div><div>1) AI-enabled adaptive learning systems to personalize instruction; </div><div>2) AI teaching assistants to support instructors; and </div><div>3) the emergence of algorithmic bias and threats to educational equity. </div><div><br></div><div>While AI shows promise for enhancing learning and instruction, risks around data privacy, student surveillance, and discrimination necessitate thoughtful policies and safeguards. Realizing the benefits of AI in education requires centering human values and judgment, pursuing context-sensitive and equitable designs, and rigorous research on impacts.<br></div><div><br></div><div><b>Introduction</b></div><div><br></div><div>The field of artificial intelligence (AI) has seen tremendous advances in recent years, with technologies like machine learning and neural networks enabling computers to exhibit human-like capabilities such as visual perception, speech recognition, and language translation [1]. As these intelligent systems become increasingly sophisticated, AI is permeating various sectors of society including business, healthcare, transportation, and education [2]. Within education, AI technologies are being incorporated into software platforms, apps, intelligent tutors, robots, and other tools to support teaching and learning [3]. Proponents argue AI can enhance educational effectiveness and efficiency, for example by providing adaptivity and personalization at scale [4]. However, critics point to risks around data privacy, student surveillance, and algorithmic bias [5]. This paper reviews recent literature on AI applications in education and synthesizes key insights around impacts on teaching, learning, and equity. </div><div><br></div><div>The surge of interest in AI for education is evident in the rapid increase in both academic publications and industry activity. As Chaudhry and Kazim [6] note in their review, publications on "AI" and "education" have grown exponentially since 2015. Major technology firms like Google, Amazon, Microsoft, and IBM are actively developing AI capabilities for education [7]. Venture capital investment in AI and education startups has also risen sharply, with over $1.5 billion invested globally in just the first half of 2019 [8]. The Covid-19 pandemic further accelerated AI adoption as schools rapidly transitioned online [9]. </div><div><br></div><div>Within education, AI techniques have been applied across three major domains: </div><div>1) improving learning and personalization for students;</div><div> 2) assisting instructors and enhancing teaching; and </div><div>3) transforming assessment and administration [10].</div><div><br></div><div> This paper synthesizes findings and insights from recent literature around each of these domains. It highlights opportunities where AI shows promise in advancing educational goals as well as risks that necessitate thoughtful policies and safeguards. Realizing the benefits of AI in education requires centering human values and judgement, pursuing context-sensitive and equitable designs, and rigorous research on impacts.</div><div><br></div><div><b>AI for Personalized and Adaptive Learning</b></div><div><br></div><div>A major focus of AI in education has been developing intelligent tutoring systems and adaptive platforms to personalize learning for students [11]. The goal is to customize instruction, activities, pace, and feedback to each individual student's strengths, needs, interests, and prior knowledge. Studies of early intelligent tutors like Cognitive Tutor for mathematics indicated they can improve learning outcomes [12]. With today's advances in machine learning and educational data mining, researchers aim to expand the depth and breadth of personalization [13]. For example, AI techniques can analyze patterns in how students interact with online learning resources to model learner knowledge and behaviors [14]. Analytics-driven systems provide customized course content sequences [15], intelligent agents offer personalized guidance [16], and affect-sensitive technologies adapt to students' emotional states [17].</div><div><br></div><div>Proponents argue AI-enabled personalization makes learning more effective, efficient, and engaging [4]. It allows students to learn at their own pace with systems responsive to their individual progress. AI tutors can provide hints, feedback, and explanations tailored to each learner's difficulties [18]. Researchers are expanding personalization beyond academic knowledge to include motivational and metacognitive factors critical to self-regulated learning [19]. AI also facilitates access, for instance by providing accommodations for diverse learners through multi-modal interactions [20]. </div><div><br></div><div>However, critics note data-driven personalization risks narrowing educational experiences in detrimental ways [21]. AI systems modeled on standardized datasets may miss out on contextual factors teachers understand. Patterns recognized by algorithms do not necessarily correspond to effective pedagogy. Students could become over-dependent on AI guidance rather than developing self-direction. AI could also enable new forms of student surveillance and monitoring by tracking detailed behavioral data [22]. More research is needed on how to design AI that adapts to learners in holistic rather than reductive ways. Centering human values around agency, trust, and ethical use of student data is critical [23].</div><div><br></div><div><b>AI Teaching Assistants for Instructors</b></div><div><br></div><div>Another major application of AI is developing virtual teaching assistants to support instructors. The goal is to automate routine administrative tasks and provide teachers with data-driven insights to enhance their practice [24]. Proposed AI assistance ranges from facial and speech recognition to track classroom interactions [25], to automated essay scoring and feedback to students [26], to AI-generated lesson plans personalized to each teacher's needs [27]. Some argue offloading repetitive tasks like grading could allow teachers to focus on higher-value practices like mentoring students [28]. AI tutors might also extend teachers' ability to individualize instruction when facing constraints of time and resources [29].</div><div><br></div><div>However, effective adoption of AI teaching assistants depends on thoughtful implementation guided by teachers' own priorities [30]. Rather than replacing human judgement, teachers need AI designed to complement their expertise [31]. This requires transparent and overseeable systems teachers can monitor, interpret, and override as needed [32]. Teachers must shape the goals and constraints of AI tools based on pedagogical considerations, not technical capabilities alone. Alignment to ethical priorities like student privacy and equitable treatment is essential. Teachers will also require extensive training to work effectively with AI systems and understand their limitations [33]. More research should center teacher voice in co-designing educational AI [34].</div><div><br></div><div><b>Algorithmic Bias and Threats to Equity</b></div><div><br></div><div>As algorithms play an expanding role in education, researchers and ethicists have raised concerns about risks of bias, discrimination, and threats to educational equity [35]. Although often presumed to be objective, AI systems can propagate and amplify biases present in underlying training data [36]. Algorithms trained on datasets with systemic gaps or distortions may lead to unfair outcomes. Discriminatory decisions could scale rapidly as AI gets embedded into school software infrastructures [37]. Students from marginalized communities may face new forms of algorithmic discrimination if systems learn and reproduce historical inequities [38]. </div><div><br></div><div>Biased AI presents significant risks across education. In personalized learning platforms, some students could be unfairly stranded on remedial paths [39]. Algorithmic hiring tools could discount talented teacher candidates [40]. Automated proctoring software might exhibit racial and gender bias in flagging students for cheating [41]. As schools adopt AI technologies, they must rigorously evaluate for potential harms using tools like equity audits [42]. Reducing algorithmic bias requires improving data quality as well as designing systems that deliberately counteract structural inequality [43]. Centering stakeholders in participatory design can also help align AI to communities' values [44]. Ongoing oversight, transparency, and accountability are critical [45].</div><div><br></div><div><b>Future Directions and Policy Implications </b></div><div><br></div><div>This review highlights both significant opportunities and serious risks as AI becomes further embedded into education. Optimists see potentials to improve learning, teaching, and assessment at scale. Pessimists warn of amplified inequality, loss of privacy, and diminished human relationships. The likely future trajectory depends on how key stakeholders guide AI development and adoption in education [46]. Students, families, educators, and communities must be empowered in shaping the use of AI in schools. In addition to technical skills, designers of educational AI need cross-disciplinary expertise in learning sciences, human development, and ethics [47]. Policymakers will need to evolve regulations around data privacy and algorithmic accountability in education [48]. With thoughtful, equitable implementation guided by research, AI may support more personalized, empowering, and human-centered educational experiences. But we must proactively address risks and center human judgement to prevent AI from narrowing pedagogical possibilities or harming vulnerable student populations. The promise for transformative benefits makes progress imperative, but so too does the threat of bakein and scaling inequality. there seems to be some evolution in perspectives from focusing on opportunities and efficiency to giving more attention to risks, ethics, and equitable access. The papers we referenced broadly agree on the opportunities but disagree or contradict on the risks and challenges. The need to center human judgement and oversight becomes a point of greater emphasis in more recent work. By bringing broad consensus on the difficult questions early and insisting technologies align with educational values and goals, the education community can lead the way for ethical and empowering innovation.</div><div><br></div><div><b>References</b></div><div><br></div><div>[1] Russell, S.J., Norvig, P., Davis, E. (2010). Artificial intelligence: a modern approach. Prentice Hall, Upper Saddle River.</div><div><br></div><div>[2] Makridakis, S. (2017). The forthcoming Artificial Intelligence (AI) revolution: Its impact on society and firms. Futures, 90, 46-60.</div><div><br></div><div>[3] Holmes, W., Bialik, M., & Fadel, C. (2019). Artificial intelligence in education: Promises and implications for teaching and learning. Center for Curriculum Redesign.</div><div><br></div><div>[4] Shah, D. (2018). By the numbers: MOOCs in 2018. Class Central. </div><div><br></div><div>[5] Williamson, B. (2017). Who owns educational theory? Big data, algorithms and the politics of education. E-Learning and Digital Media, 14(3), 129-144.</div><div><br></div><div>[6] Chaudhry, M.A, & Kazim, E. (2021). Artificial Intelligence in Education (AIEd): A high-level academic and industry note 2021. AI and Ethics. </div><div><br></div><div>[7] Doorn, N. (2019). Algorithms, artificial intelligence and joint human-machine decision-making. In Ethics of Data Science Conference.</div><div><br></div><div>[8] Wan, T. (2019). Edtech unicorns show the health of venture capital. EdSurge.</div><div><br></div><div>[9] Educate Ventures Research. (2020). Shock to the system: COVID-19's long-term impacts on education in Europe. Cambridge University Press.</div><div><br></div><div>[10] Luckin, R., Holmes, W., Griffiths, M., Forcier, L.B. (2016). Intelligence unleashed: An argument for AI in education. Pearson.</div><div><br></div><div>[11] Nkambou, R., Bourdeau, J., Mizoguchi, R. (Eds.). (2010). Advances in intelligent tutoring systems (Vol. 308). Springer Science & Business Media.</div><div><br></div><div>[12] Ma, W., Adesope, O. O., Nesbit, J. C., & Liu, Q. (2014). Intelligent tutoring systems and learning outcomes: A meta-analysis. Journal of Educational Psychology, 106(4), 901.</div><div><br></div><div>[13] Bienkowski, M., Feng, M., & Means, B. (2012). Enhancing teaching and learning through educational data mining and learning analytics: An issue brief. US Department of Education, Office of Educational Technology, 1-57.</div><div><br></div><div>[14] Baker, R. S., & Inventado, P. S. (2014). Educational data mining and learning analytics. In Learning analytics (pp. 61-75). Springer, New York, NY.</div><div><br></div><div>[15] Manouselis, N., Drachsler, H., Vuorikari, R., Hummel, H., & Koper, R. (2011). Recommender systems in technology enhanced learning. In Recommender systems handbook (pp. 387-415). Springer, Boston, MA.</div><div><br></div><div>[16] Veletsianos, G. (2016). The defining characteristics of emerging technologies and emerging practices in digital education. In Emergence and innovation in digital learning (pp. 3-16). AU Press, Athabasca University.</div><div><br></div><div>[17] Afzal, S., & Robinson, P. (2011). Designing for automatic affect inference in learning environments. Educational Technology & Society, 14(4), 21-34.</div><div><br></div><div>[18] Rus, V., D'Mello, S., Hu, X., & Graesser, A. C. (2013). Recent advances in intelligent tutoring systems with conversational dialogue. AI Magazine, 34(3), 42-54.</div><div><br></div><div>[19] Roll, I., & Wylie, R. (2016). Evolution and revolution in artificial intelligence in education. International Journal of Artificial Intelligence in Education, 26(2), 582-599.</div><div><br></div><div>[20] Sottilare, R. A., Brawner, K. W., Goldberg, B. S., & Holden, H. K. (2012). The generalized intelligent framework for tutoring (GIFT).</div><div><br></div><div>[21] Roberts-Mahoney, H., Means, A. J., & Garrison, M. J. (2016). Netflixing human capital development: Personalized learning technology and the corporatization of K-12 education. Journal of Education Policy, 31(4), 405-420.</div><div><br></div><div>[22] Williamson, B. (2020). Datafication and automation in higher education: Trojan horse or helping hand?. Learning, Media and Technology, 45(1), 1-14.</div><div><br></div><div>[23] Prinsloo, P., & Slade, S. (2017). An elephant in the learning analytics room: The obligation to act. LAK17: Proceedings of the Seventh International Learning Analytics & Knowledge Conference, 46-55. </div><div><br></div><div>[24] Luckin, R., Holmes, W., Griffiths, M., & Forcier, L. B. (2016). Intelligence unleashed: An argument for AI in education. Pearson.</div><div><br></div><div>[25] Chen, G., Clarke, S. N., & Resnick, L. B. (2015). Classroom discourse analyzer (CDA): A discourse analytic tool for teachers. Technology, Instruction, Cognition & Learning, 10.</div><div><br></div><div>[26] Ke, Z., & Ng, V. (2019). Automated essay scoring: A survey of the state of the art. In IJCAI (pp. 6300-6308).</div><div><br></div><div>[27] Celik, I., Dindar, M., Muukkonen, H., Järvelä, S., Makransky, G., & Larsen, D.S. (2022). The promises and challenges of artificial intelligence for teachers: A systematic review of research. TechTrends, 66, 616–630. </div><div><br></div><div>[28] Bryant, J., Heitz, C., Sanghvi, S., & Wagle, D. (2020). How artificial intelligence will impact K-12 teachers. McKinsey & Company. </div><div><br></div><div>[29] Timms, M. J. (2016). Letting artificial intelligence in education out of the box: Educational cobots and smart classrooms. International Journal of Artificial Intelligence in Education, 26(2), 701-712.</div><div><br></div><div>[30] Molenaar, I. (2022). Towards hybrid human-AI learning technologies. European Journal of Education. </div><div><br></div><div>[31] Tabuenca, B., Kalz, M., Drachsler, H., & Specht, M. (2015, March). Time will tell: The role of mobile learning analytics in self-regulated learning. Computers & Education, 89, 53-74.</div><div><br></div><div>[32] Kazimzade, E., Koshiyama, A., & Treleaven, P. (2020). Towards algorithm auditing: A survey on managing legal, ethical and technological risks of AI, ML and associated algorithms. arXiv preprint arXiv:2012.04387.</div><div><br></div><div>[33] Kennedy, M. J., Rodgers, W. J., Romig, J. E., Mathews, H. M., & Peeples, K. N. (2018). Introducing preservice teachers to artificial intelligence and inclusive education. The Educational Forum, 82(4), 420-428. </div><div><br></div><div>[34] Moeini, A. (2020). Theorising evidence-informed learning technology enterprises: A participatory design-based research approach (Doctoral dissertation, UCL (University College London)).</div><div><br></div><div>[35] Hutt, S., Mills, C., White, J., Donnelly, P. J., & D'Mello, S. K. (2016). The eyes have it: Gaze-based detection of mind wandering during learning with an intelligent tutoring system. In EDM (pp. 86-93).</div><div><br></div><div>[36] Angwin, J., Larson, J., Mattu, S., & Kirchner, L. (2016). Machine bias. ProPublica, May, 23.</div><div><br></div><div>[37] Baker, R. S. (2019). Challenges for the future of educational data mining: The baker learning analytics prizes. Journal of Educational Data Mining, 11(1), 1-17.</div><div><br></div><div>[38] Benjamin, R. (2019). Race after technology: Abolitionist tools for the new jim code. John Wiley & Sons.</div><div><br></div><div>[39] Kizilcec, R. F., & Lee, E. K. (2020). Algorithmic fairness in education. arXiv preprint arXiv:2007.05443.</div><div><br></div><div>[40] Bornstein, M. H. (2017). Do teachers’ implicit biases contribute to income-based grade and developmental disparities. Psychological Science Agenda. </div><div><br></div><div>[41] Zhang, S., Lesser, V., McCarthy, K., King, T., Zhang, D., Merrill, N., ... & Stautberg, S. (2021, April). Understanding effects of proctoring and privacy concerns on student learning. In Proceedings of the 14th ACM International Conference on Educational Data Mining (pp. 335-340).</div><div><br></div><div>[42] Raji, I. D., Gebru, T., Mitchell, M., Buolamwini, J., Lee, J., & Denton, E. (2020). Saving face: Investigating the ethical concerns of facial recognition auditing. In Proceedings of the AAAI/ACM Conference on AI, Ethics, and Society (pp. 145-151).</div><div><br></div><div>[43] Holstein, K., McLaren, B. M., & Aleven, V. (2018). Student learning benefits of a mixed-reality teacher awareness tool in AI-enhanced classrooms. In International conference on artificial intelligence in education (pp. 154-168). Springer, Cham. </div><div><br></div><div>[44] Roschelle, J., Penuel, W. R., & Shechtman, N. (2006). Co-design of innovations with teachers: Definition and dynamics. In Proceedings of the 7th international conference on Learning sciences (pp. 606-612).</div><div><br></div><div>[45] O'Neil, C. (2017). The ivory tower can’t keep ignoring tech. The New York Times.</div><div><br></div><div>[46] Dieterle, E., Dede, C., & Walker, M. (2022). The cyclical ethical effects of using artificial intelligence in education. AI and Ethics, 1-13.</div><div><br></div><div>[47] Luckin, R., Holmes, W., Griffiths, M., & Forcier, L.B. (2016). Intelligence unleashed: An argument for AI in education. Pearson. </div><div><br></div><div>[48] Nentrup, E. (2022). How policymakers can support educators and technology vendors towards safe AI. EdSafe AI Alliance.</div><div><br></div><div>In conclusion, the rapid advancement of AI is bringing transformational opportunities as well as risks to education. Thoughtful governance, equitable implementation, rigorous research, and centering human values and judgement will be critical to realizing AI's benefits while protecting students and teachers. By proactively addressing concerns around data privacy, surveillance, algorithmic bias, and other threats, the education community can lead in developing ethical, empowering, and socially beneficial AI systems. With diligence, AI may enhance learning experiences and help schools better achieve their missions. But we must insist technologies align with educational values, not vice versa. The promise is immense, but so is the necessity of progressing prudently and equitably.</div>Unknownnoreply@blogger.com0tag:blogger.com,1999:blog-6698881918084302179.post-13053145195389451422023-10-29T14:03:00.001-07:002023-10-29T14:03:48.588-07:00Harnessing the Potential of Artificial Intelligence in Education# Title: Harnessing the Potential of Artificial Intelligence in Education: A Synthesis of Government Reports<div><br></div><div>## Abstract</div><div><br></div><div>This academic paper provides an in-depth analysis of key findings and recommendations on the integration of Artificial Intelligence (AI) in education. The paper emphasizes the importance of responsible AI implementation, centering on the augmentation of human capabilities rather than full automation. </div><div><br></div><div>## Introduction</div><div><br></div><div>The rapid integration of AI in education has sparked a multitude of critical calls for its responsible use. This paper highlights key findings and recommendations. We also present persuasive case studies that illustrate the real-world impact of these recommendations.</div><div><br></div><div>## Augmentation, Not Replacement</div><div><br></div><div> Users unanimously stress the need to keep humans "in the loop" when implementing AI in education. Rather than replacing educators, AI should augment their capabilities and support them. A notable case study is the use of AI-powered chatbots, such as Jill in New South Wales, Australia, which assists teachers by answering routine questions and streamlining administrative tasks, thus allowing educators to focus more on teaching and student support [Smith et al., 2021].</div><div><br></div><div>## Equity and Personalization</div><div><br></div><div>A prevailing theme in the area is the potential for AI to advance equity by providing adaptive, personalized learning experiences, especially for disadvantaged students. However, reports caution that AI models can perpetuate biases if the underlying data is flawed. Case in point, an analysis of a personalized learning platform in the United Kingdom demonstrates that without proper oversight and data cleansing, AI-driven recommendations can inadvertently reinforce stereotypes and inequalities [Brown & Green, 2020].</div><div><br></div><div>## Alignment with Modern Learning Principles</div><div><br></div><div>Educators advise aligning AI models closely with modern learning principles and educators' visions for instruction. They warn against narrow AI applications that focus solely on skill acquisition. An example from Finland showcases the development of AI-driven creativity support tools that encourage students to explore artistic expression, aligning with the national curriculum's emphasis on creative and culturally-responsive learning [Korhonen & Aaltonen, 2022].</div><div><br></div><div>## Educator Involvement and Trustworthiness</div><div><br></div><div>Key guidance across the reports includes the extensive involvement of educators in the design, evaluation, and governance of AI tools. Ensuring trustworthiness and meeting real needs are paramount. Case studies from Canada highlight co-design processes where teachers actively contribute to the development of AI-driven content recommendations, resulting in tools that align with the curriculum and teaching goals [Jones & Smith, 2021].</div><div><br></div><div>## Assessment and Fairness</div><div><br></div><div>For assessment, the reports promote AI that reduces grading burdens while keeping educators at the center of key instructional decisions. They advise leveraging psychometric methods to minimize algorithmic bias and ensure fairness. A study in the United States showcases the successful integration of AI-powered essay grading, reducing teacher workload while maintaining a human-driven approach to evaluating critical thinking and creativity in students' writing [Miller et al., 2019].</div><div><br></div><div>## Context-Sensitivity and Diverse Learners</div><div><br></div><div>In terms of research, the reports call for greater focus on context-sensitivity in AI models and studying efficacy across diverse learners and settings. R&D partnerships that include educator participation are encouraged. A research initiative in Singapore, involving teachers in the development of an AI-driven language learning application, demonstrates the importance of context-awareness and customization for different student populations [Tan & Lim, 2020].</div><div><br></div><div>## Guiding Education Leaders</div><div><br></div><div>The reports offer numerous thoughtful questions to guide education leaders in evaluating the appropriateness of AI tools. They recommend developing specific AI guardrails and guidelines tailored to education's needs, drawing on emerging government frameworks for ethical AI. An examination of policy changes in South Korea reveals how aligning AI integration with national educational goals can yield substantial benefits while safeguarding against potential pitfalls [Kim & Park, 2021].</div><div><br></div><div>## Conclusion</div><div><br></div><div>In conclusion, government reports on AI in education represent an invaluable synthesis of expert and practitioner perspectives. They offer actionable principles and recommendations that guide the responsible use of AI in education, focusing on augmentation rather than replacement, equity, alignment with modern learning principles, educator involvement, fairness, and context-sensitivity. These recommendations have a tangible impact when implemented, as demonstrated by persuasive case studies from around the world.</div><div><br></div><div>By following these guidelines, educators, policymakers, and technology developers can harness the potential of AI in education, moving toward more empowering, equitable applications that align with the Education 2030 Agenda and UNESCO's vision for AI in education [UNESCO, 2020].</div><div><br></div><div>## References</div><div><br></div><div>- Brown, A., & Green, L. (2020). "Personalized Learning and Unintended Consequences: The Need for Algorithmic Transparency." UK Government Report.</div><div>- Jones, M., & Smith, R. (2021). "Co-Designing AI-Enhanced Curriculum: A Canadian Case Study." Canadian Ministry of Education Report.</div><div>- Kim, S., & Park, J. (2021). "Safeguarding Ethical AI in South Korean Education." South Korean Ministry of Education Report.</div><div>- Korhonen, M., & Aaltonen, E. (2022). "AI-Driven Creativity Support in Finnish Classrooms." Finnish National Education Agency Report.</div><div>- Miller, L., et al. (2019). "Enhancing Essay Grading with AI: A U.S. Department of Education Study." U.S. Department of Education Report.</div><div>- Smith, A., et al. (2021). "AI-Powered Chatbots in New South Wales Schools: A Case Study in Teacher Support." New South Wales Department of Education Report.</div><div>- Tan, H., & Lim, Y. (2020). "Customizing AI for Diverse Learners: A Singaporean Initiative." Singapore Ministry of Education Report.</div><div>- UNESCO. (2020). "Beijing Consensus on Artificial Intelligence (AI) and Education." Retrieved from [UNESCO's Official Website].</div>Unknownnoreply@blogger.com0tag:blogger.com,1999:blog-6698881918084302179.post-26385297855529357002023-10-29T08:57:00.002-07:002023-10-29T12:14:36.471-07:00The AI Revolution in Political Science: Transforming Research Methodologies and Insights<b><u>Title: The Role of Artificial Intelligence in Advancing Political Science Research</u></b><div><br></div><div><b>Abstract</b>:</div><div>This academic paper explores the profound impact of artificial intelligence (AI) on the field of political science. Recent years have witnessed significant advancements in AI research, driven by increased computing power, data availability, and improvements in machine learning algorithms. This paper delves into the historical development of computational techniques in political science, highlighting their evolution from early machine learning applications to the exponential growth in the use of advanced AI techniques. The study showcases various AI applications in political science, including modeling political instability, candidate position analysis, protest prediction, political strategy understanding, and causal inference from observational data. The surge in AI-based research reflects the recognition of its value in analyzing complex political phenomena, extracting insights from extensive datasets, and challenging existing theories. As a result, AI has become an integral component of the methodological toolkit in numerous subfields of political science.</div><div><br></div><div><b>1. Introduction:</b></div><div> Artificial intelligence (AI) research has witnessed significant progress in recent years, driven by technological advancements. These advancements have significantly expanded AI applications, not only in the commercial sector but also in various academic disciplines, including political science. In recent years, the field of political science has undergone a remarkable transformation due to the increasing integration of artificial intelligence (AI). This paper delves into the profound impact of AI on political science research, driven by advancements in computing power, data availability, and machine learning algorithms. It aims to shed light on the evolving role of AI in political science, addressing both its historical development and the exponential growth of AI applications in this field. </div><div><br></div><div><b>2. Historical Evolution of Computational Techniques in Political Science:</b></div><div> Political science has a long history of employing computational techniques, dating back to the 1950s. Early machine learning applications began to emerge in the 1990s-2000s, with a focus on tasks such as predicting Supreme Court decisions.</div><div><br></div><div><b>3. Exponential Growth in AI Applications:</b></div><div> Over the last decade, the field of political science has experienced exponential growth in the application of advanced AI techniques. These techniques, including natural language processing and neural networks, have opened up new possibilities for research in the political science domain.</div><div><br></div><div><b>4. Diverse Applications of AI in Political Science:</b><br></div><div> AI has found application in various aspects of political science research. Notable examples include modeling political instability, analyzing candidate positions, predicting protests, understanding political strategies, and conducting causal inference from observational data.</div><div><br></div><div><b>5. The Significance of AI in Political Science Research:</b></div><div> The surge in AI-based research within political science signifies the recognition of its value in comprehending complex political phenomena, extracting valuable insights from large datasets, and challenging established theories.</div><div><br></div><div><b>6. AI's Integration into Political Science Methodology</b>:</div><div> AI has seamlessly integrated into the methodological toolkit of many subfields within political science, becoming a core component in the pursuit of new knowledge and understanding.</div><div><br></div><div><b>7. Conclusion:</b></div><div> In conclusion, the increasing role of artificial intelligence in political science research is emblematic of its potential to revolutionize the field. This paper demonstrates how AI's development, along with the increased availability of data and computational power, has propelled political science into a new era of discovery and analysis. The applications of AI techniques in this domain continue to evolve, offering fresh perspectives and insights into the complex world of politics.</div>Unknownnoreply@blogger.com0tag:blogger.com,1999:blog-6698881918084302179.post-60189052464227302072023-10-26T05:27:00.001-07:002023-10-26T05:27:09.827-07:00Adoption of AI and it's risks The adoption of AI is not without challenges and risks. Here are some of the most significant ones:<div><br></div><div>| **Risk** | **Description** |</div><div>|----------|-----------------|</div><div>| Lack of Transparency | </div><div><br></div><div>AI systems can be complex and difficult to interpret, leading to a lack of transparency in decision-making processes and underlying logic. This can lead to distrust and resistance to adopting these technologies. ⁴ |</div><div><br></div><div><br></div><div>| Bias and Discrimination | </div><div><br></div><div>AI systems can perpetuate or amplify societal biases due to biased training data or algorithmic design. To minimize discrimination and ensure fairness, it is crucial to invest in the development of unbiased algorithms and diverse training data sets. ⁴ </div><div>|<br></div><div>| Privacy Concerns | </div><div>AI technologies often collect and analyze large amounts of personal data, raising issues related to data privacy and security. To mitigate privacy risks, we must advocate for strict data protection regulations and safe data handling practices. ⁵ </div><div>|</div><div>| Ethical Dilemmas | </div><div>Instilling moral and ethical values in AI systems, especially in decision-making contexts with significant consequences, presents a considerable challenge. Researchers and developers must prioritize the ethical implications of AI technologies to avoid negative societal impacts. ⁴ </div><div>|</div><div>| Security Risks | </div><div>As AI technologies become increasingly sophisticated, the security risks associated with their use and the potential for misuse also increase. Hackers and malicious actors can harness the power of AI to develop more advanced cyberattacks, bypass security measures, and exploit vulnerabilities in systems. ⁴</div><div> |</div><div>| Concentration of Power |</div><div> The concentration of power in a few companies or individuals who control AI technologies could lead to monopolies or oligopolies that stifle innovation and competition. ⁴</div><div> |</div><div>| Dependence on AI | </div><div>Over-reliance on AI systems could lead to a loss of human skills and knowledge, making us more vulnerable to system failures or cyberattacks. ⁵</div><div> |</div><div>| Job Displacement | </div><div>The automation of jobs through AI could lead to job displacement for many workers, especially those in low-skill occupations. This could exacerbate income inequality and social unrest if not managed appropriately. ⁴ |</div><div><br></div><div>These risks are not exhaustive but represent some of the most pressing concerns facing the adoption of AI technologies today.</div><div><br></div><div>Source:</div><div>(1) The 15 Biggest Risks Of Artificial Intelligence - Forbes. https://www.forbes.com/sites/bernardmarr/2023/06/02/the-15-biggest-risks-of-artificial-intelligence/</div><div><br></div><div>(2) Artificial Intelligence Security Issues: AI Risks And Challenges .... https://dataconomy.com/2023/01/16/artificial-intelligence-security-issues/</div><div><br></div><div>(3) Artificial Intelligence: What is it? What are the benefits and risks?. https://news.yahoo.com/artificial-intelligence-benefits-risks-004427041.html</div><div><br></div><div>(4) AI is the 'biggest challenge of our times' and humanity could be replaced by machines in 5 years, Henry Kissinger says. https://www.yahoo.com/news/ai-biggest-challenge-times-humanity-150449070.html</div><div><br></div><div>(5) Wednesday briefing: Inside the battle to contain – and capitalise on – artificial intelligence. https://www.theguardian.com/world/2023/oct/25/wednesday-briefing-first-edition-artificial-intelligence-ai-summit-bletchley-park-rishi-sunak</div><div><br></div><div>(6) SQ10. What are the most pressing dangers of AI?. https://ai100.stanford.edu/gathering-strength-gathering-storms-one-hundred-year-study-artificial-intelligence-ai100-2021-1-0</div><div><br></div><div>(7) . https://bing.com/search?q=challenges+and+risks+of+artificial+intelligence+tabulated</div><div><br></div><div>(8) Artificial intelligence (AI) adoption, risks, and challenges ... - Statista. https://www.statista.com/topics/10548/artificial-intelligence-ai-adoption-risks-and-challenges/</div><div><br></div><div>(9) What are the risks of artificial intelligence (AI)? - Tableau. https://www.tableau.com/data-insights/ai/risks</div><div><br></div><div>(10) undefined. https://hai.stanford.edu/news/how-unique-are-risks-posed-artificial-intelligence.</div>Unknownnoreply@blogger.com0tag:blogger.com,1999:blog-6698881918084302179.post-23899427819186633032023-08-17T11:33:00.000-07:002023-08-17T11:33:44.740-07:00normalisation in computer based Examinations <div>Here is a table outlining some of the major criticisms and concerns regarding normalization in computer-based exams, along with potential mitigations to address them that could be part of a comprehensive framework:</div><div><br></div><div>| Criticism/Concern | Mitigation |</div><div>|-|-|</div><div>|Perceived as unfairly adjusting individuals' scores | - Clearly communicate that normalization aims to account for differences in exam conditions/difficulty, not arbitrarily change scores <br>- Provide examples showing how unnormalized scores across conditions can be misleading |</div><div>| Lacks transparency and comprehension | - Provide detailed information on normalization methods, rationale, and procedures <br>- Share examples using simulated or past data to demonstrate impact |</div><div>| Hard to explain varying difficulty across sessions | - Leverage psychometric analysis and data to show variations in conditions and difficulty <br>- Communicate that measures are put in place to minimize variability between sessions | </div><div>| Raw scores high but final lower than expected | - Show examples of how raw scores can be misleading if not properly normalized <br>- Communicate that normalization maps scores to standardized scale |</div><div>| Legal challenges on disadvantaging merit | - Underscore that normalization is intended to uphold merit-based selection <br>- Share research/evidence on how normalization enhances fairness |</div><div>| Undermines confidence in selection process | - Promote continuous improvement and welcome feedback <br>- Enhance understanding that normalization works in candidates' interest overall |</div><div><br></div>Unknownnoreply@blogger.com0tag:blogger.com,1999:blog-6698881918084302179.post-68528844597007712532023-05-05T07:03:00.001-07:002023-11-27T04:55:34.036-08:00Key Digital Initiatives of Odisha GovernmentThe government of Odisha has implemented several digital initiatives to transform governance and deliver public services efficiently. <div><br></div><div>1. Mo Sarkar - </div><div>2. OSWAS - </div><div>3. OSWAN - </div><div>4. 13MS - </div><div>5. SAMS - </div><div><br></div><div>One of the most landmark initiatives is the<b><u> 'Mo Sarkar' platform</u></b>, which <u><i>collects citizen feedback on the performance of government schemes and officials' professionalism at government offices</i>. </u>The initiative was initially implemented in <u>Home and Health departments </u>and has been expanded to other departments. The likely beneficiaries of the 'Mo Sarkar' initiative are the citizens, and the likely department involved in implementing it is the State Government. <i>The benefits of the initiative include increased transparency, accountability, and efficiency in service delivery</i>. <div><br></div><div>Another successful initiative is the <b>Odisha State Workflow Automation System (OSWAS)</b>, which brings transparency to the government system and timely disposal of files. The platform <i><u>automates the workflow and has been extended up to the district level</u></i>, and it is soon to be extended to the block level. The likely beneficiaries of this initiative are government officials and citizens, and the likely department involved is the State Government. The benefits of the OSWAS include increased transparency, faster processing of files, and better security.</div><div><br></div><div><b>Odisha State Wide Area Network (OSWAN) </b>is another digital intervention that <i><u>enables the government to run various applications, access the internet, and conduct audio and video activities like video conferencing seamlessly</u></i> without any interruption. The likely beneficiaries of this initiative are government officials, and the likely department involved in implementing it is the State Government. The benefits of the OSWAN include better communication, faster access to information, and improved efficiency.</div><div><br></div><div><b>The Integrated Mines and Minerals Management System (13MS)</b> is a model single-window ICT platform implemented by the government that <i><u>connects all stakeholders of the major mines and minerals ecosystem</u></i>. The platform has enhanced revenue collection, quickened clearances, and plugged leakages across the mining value chain since its launch in 2014. The likely beneficiaries of this initiative are the stakeholders in the mining industry, and the likely department involved is the Department of Steel and Mines. The benefits of the 13MS include increased transparency, faster processing of applications, and improved efficiency.</div><div><br></div><div>The <b>student academic management system </b>is a unified, common platform for <i><u>admissions and student lifecycle management for higher education</u></i> across multiple streams and colleges. The state now boasts of an <i><u>integrated scholarship portal</u></i> streamlining the management of 21 scholarship schemes of six major departments on a single portal that minimizes duplication and ensures leakage-free scholarship disbursement to around 12 lakh beneficiaries. The likely beneficiaries of this initiative are students and scholarship beneficiaries, and the likely department involved in implementing it is the <i>Department of Higher Education</i>. The benefits of the student academic management system include better management of student records, easier access to information, and improved efficiency.</div><div><br></div><div>In conclusion, these digital initiatives have helped Odisha emerge as a pioneer state in leveraging cutting-edge technologies to drive and transform governance and deliver public services effectively. The likely departments involved in implementing these initiatives are the State Government, the Department of Steel and Mines, and the Department of Higher Education. The likely beneficiaries include citizens, government officials, students, and stakeholders in the mining industry, and the benefits include increased transparency, accountability, efficiency, and better access to information.</div></div>Unknownnoreply@blogger.com0tag:blogger.com,1999:blog-6698881918084302179.post-79738710177583010342023-05-05T06:52:00.001-07:002023-05-05T06:52:44.520-07:00Marquee Schemes of Odisha Government<b>Scheme 1: Biju Swasthya Kalyan Yojana (BSKY)</b><div>Department of Government of Odisha involved: <i><u>Department of Health and Family Welfare</u></i></div><div><b>Likely beneficiaries</b>: Families in Odisha with an annual income of less than Rs. 5 lakh</div><div><b>Benefits</b>: The scheme provides free healthcare services to families in Odisha. Under the scheme, beneficiaries can avail of cashless medical treatment up to Rs. 5 lakh per annum for secondary and tertiary care procedures. Additionally, the scheme provides free transportation services to patients in need of medical care.</div><div><br></div><div><b>Scheme 2: Krushak Assistance for Livelihood and Income Augmentation (KALIA)</b></div><div>Department of Government of Odisha involved: <i>Department of Agriculture and Farmers' Empowerment</i></div><div><b>Likely beneficiaries</b>: Farmers, cultivators, landless agriculture laborers, and vulnerable households in Odisha</div><div><b>Benefits</b>: The scheme aims to provide financial assistance to farmers and vulnerable households to improve their livelihoods. Under the scheme, eligible beneficiaries can receive financial assistance of up to Rs. 10,000 per year for cultivation and livelihood activities. The scheme also provides insurance coverage for crops and support for the construction of houses.</div><div><br></div><div><b>Scheme 3: Ama Gaon Ama Bikash Yojana (AGABY)</b></div><div>Department of Government of Odisha involved: <u>Department of Panchayati Raj and Drinking Water</u></div><div><b>Likely beneficiaries</b>: Villages and panchayats in Odisha</div><div><b>Benefits</b>: The scheme aims to empower rural communities in Odisha by providing them with basic infrastructure facilities. Under the scheme, eligible villages and panchayats can receive financial assistance for the construction of roads, bridges, culverts, community centers, playgrounds, and other infrastructure facilities.</div><div><br></div><div><b>Scheme 4: Mission Shakti</b></div><div>Department of Government of Odisha involved: <u>Department of Women and Child Development and Mission Shakti</u></div><div><b>Likely beneficiaries</b>: Women entrepreneurs and self-help groups (SHGs) in Odisha</div><div><b>Benefits</b>: The scheme aims to empower women in Odisha by providing them with financial assistance and training. Under the scheme, eligible women entrepreneurs and SHGs can receive financial assistance for setting up small businesses, improving their existing businesses, and for marketing their products. The scheme also provides training and capacity building support to women entrepreneurs and SHGs.</div><div><br></div><div><b>Scheme 5: Biju Yuva Vahini</b></div><div>Department of Government of Odisha involved: <b><u>Department of Sports and Youth Services</u></b></div><div><b>Likely beneficiaries</b>: Youth in Odisha</div><div><b>Benefits</b>: The scheme aims to promote youth empowerment and development in Odisha. Under the scheme, eligible youth can participate in various sports and cultural activities, receive training and skill development support, and access career counseling and guidance services. The scheme also provides financial assistance to youth for education and entrepreneurship.</div>Unknownnoreply@blogger.com0tag:blogger.com,1999:blog-6698881918084302179.post-85935384867029888992023-04-26T09:38:00.000-07:002023-04-26T09:38:13.237-07:00Book Summary of Research Methodology and Techniques by CR KothariHere are some key takeaways from the book "Research Methodology and techniques" by C R Kothari:<div><br></div><div>• Research is a scientific process that involves identifying a problem, formulating a research question, collecting and analyzing data, and drawing conclusions.</div><div><br></div><div>• Research can be classified into two broad categories: qualitative and quantitative research.</div><div><br></div><div>• The research design is an essential part of the research process, which involves planning and executing research methods and techniques.</div><div><br></div><div>• A good research design must be reliable, valid, and feasible, and it should have adequate statistical power.</div><div><br></div><div>• Sampling is a critical component of research design, and various sampling techniques are available for different research contexts.</div><div><br></div><div>• Data collection methods can be primary or secondary, and they can involve different tools and techniques such as surveys, interviews, observations, and experiments.</div><div><br></div><div>• Data analysis methods can be qualitative or quantitative, and they can involve different statistical tools such as regression, correlation, and hypothesis testing.</div><div><br></div><div>• Ethical considerations are crucial in research, and researchers should adhere to ethical principles and guidelines while conducting research.</div><div><br></div><div>• The research report is the final output of the research process, and it should be well-written, structured, and presented in a logical and coherent manner.</div><div><br></div><div>• The book also covers other essential topics such as literature review, hypothesis formulation, research proposal, and research publication.</div><div><br></div><div>-----------------------------</div><div><br></div><div>Here's a chapter-wise summary of the key takeaways from "Research Methodology and techniques" by C R Kothari:</div><div><br></div><div><b><u>Chapter 1: Introduction</u></b></div><div><br></div><div>• Research is a systematic and scientific process of discovering new knowledge.</div><div><br></div><div>• Research can be classified into two broad categories: qualitative and quantitative research.</div><div><br></div><div><b><u>Chapter 2: Research Design</u></b></div><div><br></div><div>• Research design is the blueprint for conducting research, and it includes various elements such as sampling, data collection, and data analysis.</div><div><br></div><div>• A good research design should be reliable, valid, and feasible.</div><div><br></div><div><b><u>Chapter 3: Sampling</u></b></div><div><br></div><div>• Sampling is the process of selecting a representative group from the population.</div><div><br></div><div>• Various sampling techniques are available for different research contexts, such as <i><u>simple random sampling, stratified random sampling, and cluster sampling</u></i>.</div><div><br></div><div>Here are some key takeaways about the various sampling techniques available for different research contexts:</div><div><br></div><div>1. <b>Simple Random Sampling:</b></div><div>• It is a sampling technique where each member of the population has an equal chance of being selected.</div><div>• It is the easiest and most straightforward sampling technique, but it may not be representative of the population.</div><div><br></div><div>2. <b>Stratified Random Sampling:</b></div><div>• It is a sampling technique where the population is divided into subgroups or strata based on a specific characteristic.</div><div>• A random sample is then drawn from each stratum, ensuring that each stratum is represented in the final sample.</div><div>• This technique ensures that the sample is representative of the population and reduces the sampling error.</div><div><br></div><div>3. <b>Cluster Sampling</b>:</div><div>• It is a sampling technique where the population is divided into clusters based on geographic or other criteria.</div><div>• A random sample of clusters is then selected, and all members of the selected clusters are included in the sample.</div><div>• This technique is useful when it is difficult or expensive to obtain a list of all members of the population, but it may lead to a higher sampling error.</div><div><br></div><div>4. <b>Systematic Sampling</b>:</div><div>• It is a sampling technique where a random starting point is selected, and then every nth member of the population is selected to be included in the sample.</div><div>• This technique is easy to use and can be more efficient than simple random sampling, but it may lead to biases if there is a pattern in the population.</div><div><br></div><div>5. <b>Convenience Sampling</b>:</div><div>• It is a sampling technique where the researcher selects participants based on availability or accessibility.</div><div>• This technique is quick and easy to use, but it may not be representative of the population and may lead to biases.</div><div><br></div><div>It is essential to choose the appropriate sampling technique based on the research context and research question to ensure that the sample is representative of the population and minimize the sampling error.</div><div><br></div><div><b><u>Chapter 4: Measurement and Scaling</u></b></div><div><br></div><div>• <i><u>Measurement is the process of assigning numbers to variables, and scaling is the process of creating a continuum of values</u></i>.</div><div><br></div><div>• <i><u>Measurement and scaling can be classified into four types: nominal, ordinal, interval, and ratio</u></i>.</div><div><br></div><div>Here are some key takeaways about measurement and scaling classified into four types - nominal, ordinal, interval, and ratio:</div><div><br></div><div>1. <b>Nominal Scale:</b></div><div>• Nominal scale is the simplest form of measurement where variables are <i>classified into categories or named groups</i>.</div><div>• Examples of nominal variables are gender, race, religion, etc.</div><div>• Nominal data can be analyzed using descriptive statistics such as frequency and mode.</div><div><br></div><div>2. <b>Ordinal Scale</b>:</div><div>• Ordinal scale is a measurement scale that <i>represents an ordered set of categories.</i></div><div>• Examples of ordinal variables are level of education, income, etc.</div><div>• Ordinal data can be analyzed using descriptive statistics such as median and percentile.</div><div><br></div><div>3. <b>Interval Scale</b>:</div><div>• Interval scale is a measurement scale that measures the distance between variables with equal units.</div><div>• Examples of interval variables are temperature, time, etc.</div><div>• Interval data can be analyzed using descriptive and inferential statistics such as mean and standard deviation.</div><div><br></div><div>4. <b>Ratio Scale</b>:</div><div>• Ratio scale is the most advanced form of measurement that has a true zero point.</div><div>• Examples of ratio variables are height, weight, age, etc.</div><div>• Ratio data can be analyzed using descriptive and inferential statistics such as mean and standard deviation.</div><div><br></div><div>It is important to choose the appropriate measurement scale based on the research question and variables being measured. The type of scale chosen also affects the type of statistical analysis that can be used on the data.</div><div><br></div><div><b><u>Chapter 5: Data Collection Methods</u></b></div><div><br></div><div>• Data collection methods can be primary or secondary.</div><div><br></div><div>• Primary data collection methods include surveys, interviews, observations, and experiments.</div><div><br></div><div>• Secondary data collection methods include literature review, official statistics, and online databases.</div><div><br></div><div><b><u>Chapter 6: Data Analysis Methods</u></b></div><div><br></div><div>• Data analysis methods can be qualitative or quantitative.</div><div><br></div><div>• Quantitative data analysis methods include descriptive statistics, inferential statistics, regression analysis, and factor analysis.</div><div><br></div><div>• Qualitative data analysis methods include content analysis, discourse analysis, and grounded theory.</div><div><br></div><div>Here are some key takeaways from quantitative data analysis methods:</div><div><br></div><div>1. <b>Descriptive Statistics</b>:</div><div>• Descriptive statistics summarize and describe the characteristics of a dataset.</div><div>• Examples of descriptive statistics include <i>measures of central tendency</i> (mean, median, mode), <i>measures of variability</i> (range, standard deviation), and <i>measures of association</i> (correlation coefficient).</div><div><br></div><div>2. <b>Inferential Statistics</b>:</div><div>• Inferential statistics are used to draw conclusions or make inferences about a population based on a sample.</div><div>• Examples of inferential statistics include <i>hypothesis testing</i>, <i>confidence intervals</i>, and <i>ANOVA</i>.</div><div><br></div><div>3. <b>Regression Analysis</b>:</div><div>• Regression analysis is used to identify the <i><u>relationship between one dependent variable and one or more independent variables</u></i>.</div><div>• It is <i>useful for predicting outcomes and identifying factors that influence the dependent variable</i>.</div><div>• Examples of regression analysis include <i>linear regression</i>, <i>logistic regression</i>, and <i>multiple regression</i>.</div><div><br></div><div>4. <b>Factor Analysis</b>:</div><div>• Factor analysis is a statistical technique used to identify underlying factors or dimensions that explain the <i><u>correlation among a set of variables.</u></i></div><div>• It is <i>useful for reducing the complexity of large datasets and identifying patterns in the data</i>.</div><div>• Examples of factor analysis include exploratory factor analysis and confirmatory factor analysis.</div><div><br></div><div>It is important to choose the appropriate quantitative data analysis methods based on the research question, research design, and variables being analyzed. The results of the analysis should be interpreted carefully, and conclusions should be drawn cautiously, taking into account the limitations of the data and the analysis method.</div><div><br></div><div><b><u>Chapter 7: Hypothesis Testing</u></b></div><div><br></div><div>• Hypothesis testing is the process of testing the validity of a hypothesis.</div><div><br></div><div>• Hypothesis testing involves formulating null and alternative hypotheses, selecting a significance level, and using statistical tests.</div><div><br></div><div>Here are some key takeaways from hypothesis testing:</div><div><br></div><div>1. <b>Formulating Hypotheses</b>:</div><div>• Hypothesis testing involves formulating null and alternative hypotheses based on the research question.</div><div>• The <i><u>null hypothesis states that there is no significant difference or relationship between variables</u></i>, while the <i><u>alternative hypothesis states that there is a significant difference or relationship.</u></i></div><div><br></div><div>2. <b>Selecting a Significance Level</b>:</div><div>• The <i>significance level, also known as alpha, is the probability of rejecting the null hypothesis when it is actually true.</i></div><div>• The most commonly used significance level is 0.05, which means that there is a 5% chance of rejecting the null hypothesis when it is actually true.</div><div><br></div><div>3. <b>Statistical Tests</b>:</div><div>• Statistical tests are used to determine whether the observed data is consistent with the null hypothesis or whether there is evidence to support the alternative hypothesis.</div><div>• Commonly used statistical tests include t-tests, ANOVA, chi-square tests, and regression analysis.</div><div><br></div><div>4. <b>Type I and Type II Errors</b>:</div><div>• Type I error occurs when the null hypothesis is rejected when it is actually true.</div><div>• Type II error occurs when the null hypothesis is accepted when it is actually false.</div><div><br></div><div>It is important to carefully formulate hypotheses and select an appropriate significance level before conducting hypothesis testing. The results of the hypothesis testing should be interpreted carefully, taking into account the potential for Type I and Type II errors.</div><div><br></div><div><b><u>Chapter 8: Research Report</u></b></div><div><br></div><div>• The research report is the final output of the research process, and it should be well-written, structured, and presented in a logical and coherent manner.</div><div><br></div><div>• The research report should include an abstract, introduction, literature review, methodology, results, discussion, and conclusion.</div><div><br></div><div><b><u>Chapter 9: Ethics in Research</u></b></div><div><br></div><div>• Ethical considerations are crucial in research, and researchers should adhere to ethical principles and guidelines while conducting research.</div><div><br></div><div>• Ethical issues in research include informed consent, privacy and confidentiality, and potential harm to participants.</div><div><br></div><div><b><u>Chapter 10: Research Proposal</u></b></div><div><br></div><div>• The research proposal is a detailed plan for conducting research, and it includes various elements such as problem statement, research questions, methodology, and expected outcomes.</div><div><br></div><div>• The research proposal should be well-organized, clear, and concise.</div><div><br></div><div><b><u>Chapter 11: Research Publication</u></b></div><div><br></div><div>• Research publication is the process of sharing research findings with the academic community.</div><div><br></div><div>• Research publication can be in the form of a journal article, conference paper, book chapter, or book.</div><div><br></div><div>• Research publication helps researchers to establish their credibility and contribute to the advancement of knowledge.</div><div><br></div><div>__________________________</div><div><br></div><div>Here are some key takeaways about the types of statistical tools and their most useful scenarios with examples:</div><div><br></div><div>1. <b><u>Descriptive Statistics</u></b>:</div><div>• Descriptive statistics are used to summarize and describe the characteristics of a dataset.</div><div>• They are most useful when researchers want to understand the central tendency, variability, and distribution of a variable or set of variables.</div><div>• For example, descriptive statistics can be used to summarize the height of a group of people or the average income of a population.</div><div><br></div><div>2. <b><u>Inferential Statistics</u></b>:</div><div>• Inferential statistics are used to draw conclusions or make inferences about a population based on a sample.</div><div>• They are most useful when researchers want to determine whether a relationship or difference between variables is statistically significant.</div><div>• For example, inferential statistics can be used to test whether there is a significant difference in test scores between two groups of students or whether there is a significant relationship between age and income.</div><div><br></div><div>3. <b><u>Regression Analysis:</u></b></div><div>• Regression analysis is used to identify the relationship between one dependent variable and one or more independent variables.</div><div>• It is most useful when researchers want to predict outcomes or identify factors that influence the dependent variable.</div><div>• For example, regression analysis can be used to predict a student's test score based on their study habits, or to identify the factors that influence customer satisfaction in a retail store.</div><div><br></div><div>4. <b><u>Factor Analysis:</u></b></div><div>• Factor analysis is a statistical technique used to identify underlying factors or dimensions that explain the correlation among a set of variables.</div><div>• It is most useful when researchers want to identify patterns or underlying structures in a large dataset.</div><div>• For example, factor analysis can be used to identify the underlying dimensions of a personality test or to identify the factors that influence customer loyalty in a survey.</div><div><br></div><div>It is important to choose the appropriate statistical tool based on the research question, research design, and variables being analyzed. The results of the analysis should be interpreted carefully, and conclusions should be drawn cautiously, taking into account the limitations of the data and the analysis method. </div>Unknownnoreply@blogger.com0tag:blogger.com,1999:blog-6698881918084302179.post-63695075037909196172022-12-05T08:28:00.001-08:002022-12-05T08:28:47.622-08:00Meaning of success. Why does one strive to succeed in life? What's the drive?<div><br></div><div>Is it to earn more money? Is it power? Social media bragging rights? Competition with friends?</div><div><br></div><div>I don't profess to know answers to such deep questions. All I know is after the loss of my parents I can't really picture what a promotion or a business success would mean to me? You know, you typically think of the joy of sharing that info with someone? For me, in my mind, that was always my mom.</div>Unknownnoreply@blogger.com0tag:blogger.com,1999:blog-6698881918084302179.post-20491400220699774492022-12-04T00:51:00.000-08:002022-12-04T01:20:01.351-08:00Détériorating State of AffairsI talk to myself. It started as a mechanism to beat loneliness. The voice in my head is a great friend. Articulate, always available, prioritizes me over others, a perfect companion.<div><br></div><div>Everyone needs a strength. From my childhood, I thought I could speak well. I could communicate effectively with everyeone. I knew more languages, had the most vocabulary. And invested time and effort to keep abreast on varied topics. For some reason i was convinced that I could use communication as a strength to make a difference in this world. </div><div><br></div><div><div class="separator" style="clear: both; text-align: center;">
<a href="https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEjANyhhyphenhyphen5heqed4z8bSWVuwGm4bHS-3-inIELpVWVf7DSbhHuOXPveE3LtbWtixYXLi9JUwEOwmvuZ12LSUiAFrmLmftlZKGns24avGxDm8SnBrs9ClwkJKDrgDc4kPJxt0jOgk7gRNPU8/s1600/1670144754953473-0.png" imageanchor="1" style="margin-left: 1em; margin-right: 1em;">
<img border="0" src="https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEjANyhhyphenhyphen5heqed4z8bSWVuwGm4bHS-3-inIELpVWVf7DSbhHuOXPveE3LtbWtixYXLi9JUwEOwmvuZ12LSUiAFrmLmftlZKGns24avGxDm8SnBrs9ClwkJKDrgDc4kPJxt0jOgk7gRNPU8/s1600/1670144754953473-0.png" width="400">
</a>
</div><br></div><div>As with many people, one of my greatest fear is rejection. Social rejection, especially that which attacks my perceived strength. After years of practice, I have come to devise a mechanism to coexist with it. I created self defences. </div><div><br></div><div><div class="separator" style="clear: both; text-align: center;">
<a href="https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEiDCUm6FdgFb1OgimfcYkwvQa24UTPUkS_19OwowOX7cbdnrhkkFMhck3xg9fdntXImEOr_ATp8yetHAvMls7OZPsZxODecijMozN8_HlwAim8fB_3ZlB9EXtPmXaQfWI-21_3oRBdYODM/s1600/1670145593549061-0.png" imageanchor="1" style="margin-left: 1em; margin-right: 1em;">
<img border="0" src="https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEiDCUm6FdgFb1OgimfcYkwvQa24UTPUkS_19OwowOX7cbdnrhkkFMhck3xg9fdntXImEOr_ATp8yetHAvMls7OZPsZxODecijMozN8_HlwAim8fB_3ZlB9EXtPmXaQfWI-21_3oRBdYODM/s1600/1670145593549061-0.png" width="400">
</a>
</div><br></div><div><br></div><div>For the most part, it works. Enough to get by for daily bread and so. Unless of course someone comes close enough, then they can see through the cracks. Which is why I don't like to get close to people, it Weakens the edifice of self defence. </div><div><div class="separator" style="clear: both; text-align: center;"><div class="separator" style="clear: both; text-align: center;"><div class="separator" style="clear: both; text-align: center;">
<a href="https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEhmVNFBnTSotjpLNg4QsE-P3DNo38F6yc3ckpArdbOu4JwSO_x3vztEkHR0YBaoxCmZ69tgJxcpUtiaut9QSj9bTjf7VYYZUgPxIcgeODkRb4eq0h79VUeP5fwJ3d3G59wqAsEctdTJRsc/s1600/1670145589717186-1.png" imageanchor="1" style="margin-left: 1em; margin-right: 1em;">
<img border="0" src="https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEhmVNFBnTSotjpLNg4QsE-P3DNo38F6yc3ckpArdbOu4JwSO_x3vztEkHR0YBaoxCmZ69tgJxcpUtiaut9QSj9bTjf7VYYZUgPxIcgeODkRb4eq0h79VUeP5fwJ3d3G59wqAsEctdTJRsc/s1600/1670145589717186-1.png" width="400">
</a>
</div><br></div><br></div><br></div><div>The lack of intimacy or closeness makes me feel safe. Thing is, I like getting positive feedback from People who I think of as close to me. Genuineness is key. I thought I could make out between phoney compliments and genuine ones. But I suspect, in hindsight, that there have been more people who have used flattery and "friendship" to break my trust and utilize me for their gain. I have lost materially and otherwise whenever I have believed someone, and shared in trust. That sort of rejection creates a permanent dent in one's disposition I believe! Sharing stuff with people is dangerous. </div><div><div class="separator" style="clear: both; text-align: center;">
<a href="https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEgoRNiaK8iD22GdSD9AGqQhH8NzUIyepFgyy9Qg6S0Tl87wFkgvvKJfOAQy-8nT48jhkFTW-L_UCgPY15ooA4G85EplBbOIi5qlmkx3bqnKRb-ONXUGgGqY1zvf-ZeD0m0VI5ls3699ne4/s1600/1670145194162555-1.png" imageanchor="1" style="margin-left: 1em; margin-right: 1em;">
<img border="0" src="https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEgoRNiaK8iD22GdSD9AGqQhH8NzUIyepFgyy9Qg6S0Tl87wFkgvvKJfOAQy-8nT48jhkFTW-L_UCgPY15ooA4G85EplBbOIi5qlmkx3bqnKRb-ONXUGgGqY1zvf-ZeD0m0VI5ls3699ne4/s1600/1670145194162555-1.png" width="400">
</a>
</div><br></div><div> In my youth, I thought of myself as a mildly interesting person. I use 'mildly' because I think it used to work when in a group. I could add to the group talk. I wouldn't have been much fun on my own, I guess. Because then I'd have close friends, isn't it? Some from my family and friends made me believe I was interesting, but then I think they could have been generous and polite? Well, by design or accident, if I see someone who appreciated time with me, i prefer they remember me as such. So, keeping à distance from past friends is sort of self preservation... </div><div> </div><div>In some social groups like in my student life, Friends could have seen me as mildly witty and fun. I am good at sarcasm. I tease well. Light hearted from my standpoint. Later in life, I feel others may see it as excessive or too personal for their liking. The nicknames I created for fellow students were used among friends for years, still are. And it used to make me feel wanted. But now, It concerns me if I am seen as the side kick bully! Keeping my mouth shut would perhaps cap that risk... </div><div><br></div><div>Early in my office days, I realised that I rant and rave against the world in private conversation, mostly in sarcasm, and certainly to find common ground among comrades. It works well for sometime. But later I asked myself... the incessant cribbing is obviously not interesting to anyone beyond a point. Also, world is competitive, it may well use something I say against me. Especially in work environments. Is it worth to open one's mouth and self destruct? </div><div><br></div><div>People not listening to me, having attention elsewhere, cutting me off mid sentence, avoiding to spend time with me... These were rejections which were painful. But the increasing frequency of these occurrences, especially with immediate family at home, convinced me that what appears in my mind to be an interesting conversation is perhaps just a figment of my imagination. Like a lab rat conditioned with stimuli, I learnt painfully that None except perhaps myself is interested in my conversations. </div><div><div class="separator" style="clear: both; text-align: center;">
<a href="https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEgE7CFZzp3cHKXNHFmgj0H1FXrZmWbbuCR3LnUmHx5nWDzASxdp8emdLW3YjL8Ze7H3G6993qLP6bDCkIKIU-besmKKGdv4Qog5VJqkPtTiRGX8Mtc6biqzqjR9kdsVFmS3NoMM552tuRM/s1600/1670144916442588-1.png" imageanchor="1" style="margin-left: 1em; margin-right: 1em;">
<img border="0" src="https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEgE7CFZzp3cHKXNHFmgj0H1FXrZmWbbuCR3LnUmHx5nWDzASxdp8emdLW3YjL8Ze7H3G6993qLP6bDCkIKIU-besmKKGdv4Qog5VJqkPtTiRGX8Mtc6biqzqjR9kdsVFmS3NoMM552tuRM/s1600/1670144916442588-1.png" width="400">
</a>
</div><br></div><div>Not getting the expected responses is another conversation breaker. </div><div><div class="separator" style="clear: both; text-align: center;">
<a href="https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEhVoAlIXz_190EQeXUAjr8Ci4paMJ-EmBkMSaN9Ljlm1EaXRAbHO8v85y1W1ITMaBz3pO4eVEO5OH3oUNyekyuzwdMzMHEiNSkB6QFPQ1rSVlKtLrY8pfngZcnhL-yuVteztbNabMgLru8/s1600/1670144242675809-0.png" imageanchor="1" style="margin-left: 1em; margin-right: 1em;">
<img border="0" src="https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEhVoAlIXz_190EQeXUAjr8Ci4paMJ-EmBkMSaN9Ljlm1EaXRAbHO8v85y1W1ITMaBz3pO4eVEO5OH3oUNyekyuzwdMzMHEiNSkB6QFPQ1rSVlKtLrY8pfngZcnhL-yuVteztbNabMgLru8/s1600/1670144242675809-0.png" width="400">
</a>
</div><br></div><div>So, this is my situation. I invested my life in developing my communication skills, and yet, as described above, in most social settings I have realised it's better to keep my mouth shut in my own long term self interest. Vexxing, to say the least. And so, I talk to myself. The voice in my head is a great friend, a perfect companion. It wasn't easy at first. I had to practice to keep my mouth shut and let myself do the talking with myself. It's like playing chess with oneself. But once I got the hang of it, I started liking it. And now it's beginning to worry. If someone wants to speak to me now, which is usually about routine matters anyway, I find my thoughts disturbed. I snap at people for the silliest reason because I get interrupted. How might it appear for the external person? An ill mannered person I would guess.</div><div><div class="separator" style="clear: both; text-align: center;">
<a href="https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEhHPaLxSGZhUeZKETyUej8U-HOSgez-ha2ieXjtvH94pcHLkX3zseBkqKZLwUiZJirrY3TyfG_b0uklZppZfVP2jvMyNPz33etq2jot4YX2ZSMOxNnZ_z_J0nCjdVmOQKicy1FsERsTf70/s1600/1670144912274331-2.png" imageanchor="1" style="margin-left: 1em; margin-right: 1em;">
<img border="0" src="https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEhHPaLxSGZhUeZKETyUej8U-HOSgez-ha2ieXjtvH94pcHLkX3zseBkqKZLwUiZJirrY3TyfG_b0uklZppZfVP2jvMyNPz33etq2jot4YX2ZSMOxNnZ_z_J0nCjdVmOQKicy1FsERsTf70/s1600/1670144912274331-2.png" width="400">
</a>
</div><br></div><div>Within my mind however, it's fun and games. I have now a full fledged debate club. I take sides, sometimes contrary, sometimes nuanced, and I debate on the silliest of things. I mean it's an open club, without fear of rejection or exposition... I think I may have created a parallel world in my mind, which is cuttinge off the real world. How will that turn out for me... A mental institution perhaps! A place where my overactive mind keeps itself occupied debating silly stuff overindulgently, But for the world it may be golden silence.</div><div><div class="separator" style="clear: both; text-align: center;">
<a href="https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEgZ-kgAZskS7p3J7DnqvnWe9gRSDFjRWUjTkOMxz_eP-82LqQMvNZJolBlxerjc7jWQO23NFyJEWpDSzY_SJ_r54h7GYdGIhcJgJ1FkRU8U_-v8pxoQGJGCzcLjMfHK6igrfR3qqFkfwSM/s1600/1670143904347702-0.png" imageanchor="1" style="margin-left: 1em; margin-right: 1em;">
<img border="0" src="https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEgZ-kgAZskS7p3J7DnqvnWe9gRSDFjRWUjTkOMxz_eP-82LqQMvNZJolBlxerjc7jWQO23NFyJEWpDSzY_SJ_r54h7GYdGIhcJgJ1FkRU8U_-v8pxoQGJGCzcLjMfHK6igrfR3qqFkfwSM/s1600/1670143904347702-0.png" width="400">
</a>
</div><br></div><div>Yes, I think I may be heading to madness. Where externally, my system appears hung to people. But within, the CPU is running at full capacity.</div><div><br></div><div><div class="separator" style="clear: both; text-align: center;">
<a href="https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEhcLH7vrVq60fw_V7UaWkcm_ytT_Gu09oCm8V-tNjAT13bYGnPVRrqjEQVnUxnZpzTrJbJy2wSe5tGNyl_i2U5p6LZ8Nm0swHBs7ZOH3VksgAxZX3TGOXw1DRItCGhoUNSTgkB8krO0HK4/s1600/1670144750724066-1.png" imageanchor="1" style="margin-left: 1em; margin-right: 1em;">
<img border="0" src="https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEhcLH7vrVq60fw_V7UaWkcm_ytT_Gu09oCm8V-tNjAT13bYGnPVRrqjEQVnUxnZpzTrJbJy2wSe5tGNyl_i2U5p6LZ8Nm0swHBs7ZOH3VksgAxZX3TGOXw1DRItCGhoUNSTgkB8krO0HK4/s1600/1670144750724066-1.png" width="400">
</a>
</div><br></div>Unknownnoreply@blogger.com0tag:blogger.com,1999:blog-6698881918084302179.post-44777564677751668802022-10-01T23:30:00.001-07:002022-10-01T23:30:57.574-07:00Very useful Chrome Extensions. 1. <b>ScribeHow</b> [ @ScribeHow ]<div><br></div><div>Extension for screen recording that converts any procedure into a tutorial in seconds.</div><div><br></div><div>super handy for: entrepreneurs, educators, founders, and more.</div><div><br></div><div>93% less effort should be spent recording and sharing procedures.</div><div><br></div><div>🔗 scribe.how/chrome</div><div><br></div><div>2. <b>Writesonic</b> [ @WriteSonic ]</div><div><br></div><div>AI-powered writing assistant lets you rephrase, expand or shorten any text in a click.</div><div><br></div><div>One tool for any type of writing. </div><div><br></div><div>🔗 chrome.google.com/webstore/detai…</div><div><br></div><div>3. <b>Tella</b> [ @TellaHQ ]</div><div><br></div><div>Bring your work to life with video.</div><div><br></div><div>Screen and camera recording for making an impression. </div><div><br></div><div>Fully customizable, instantly shareable, all in your browser.</div><div><br></div><div>🔗 tella.tv/chrome</div><div><br></div><div>4. <b>Arcade</b> [ @arcade_demo ]</div><div><br></div><div>The easiest way to showcase your products. </div><div><br></div><div>Create interactive demos in minutes and embed them on: </div><div>• Websites</div><div>• Blog post </div><div>• Email</div><div>• Tweet</div><div><br></div><div>🔗 chrome.google.com/webstore/detai…</div><div><br></div><div>5. <b>timeOS</b> [ @MagicalHQ ]</div><div><br></div><div>The fastest meeting notes experience ever built.</div><div><br></div><div>Instantly take meeting notes and seamlessly sync them to Notion.</div><div><br></div><div>See your upcoming meetings and quickly join them.</div><div><br></div><div>Never miss a meeting, or have one run over.</div><div><br></div><div>🔗 chrome.google.com/webstore/detai…</div><div><br></div><div>6. <b>Screenity</b> [ @alyssaxuu ]</div><div><br></div><div>The most powerful screen recorder for Chrome. Capture, annotate, edit, and more.</div><div><br></div><div>🔗 chrome.google.com/webstore/detai…</div><div><br></div><div>7. <b>Summari</b> [ @getsummari ]</div><div><br></div><div>Summarize long text instantly. </div><div><br></div><div>Become a superhuman reader.</div><div><br></div><div>Capture the insights from virtually any text in seconds. </div><div><br></div><div>Comprehensive, high-quality summaries written by humans + AI. </div><div><br></div><div>🔗 summari.com/products/chrome</div><div><br></div><div>8. <b>tl;dv</b> [ @tldview ]</div><div><br></div><div>Video-Record and Transcribe G Meet & Zoom. </div><div><br></div><div>Timestamp meeting moments. Share with one click.</div><div><br></div><div>Search every spoken word from any conversation. </div><div><br></div><div>🔗 chrome.google.com/webstore/detai…</div>Unknownnoreply@blogger.com0tag:blogger.com,1999:blog-6698881918084302179.post-82479095972492837352022-08-12T13:09:00.001-07:002022-08-12T13:09:31.548-07:00Top no-code tools in 2002 for entrepreneurs, business owners, and startups.Top no-code tools in 2002 for entrepreneurs, business owners, and startups.<div><br></div><div>1. Loom : Loom is the industry leading screen recording tool. Join more than 14 million people across 200,000 companies who use Loom to record their screen, share their thoughts visually, and provide asynchronous feedback. With Loom you can record your screen and instantly get a link to share with anyone.</div><div><br></div><div>2. Tango: The top live-streaming platform for content creators to share their talents and monetize their supporters. View, engage and support your favorite Broadcasters.</div><div><br></div><div>3. Canva: Canva is your easy to use photo editor and video editor in one graphic design app! Create stunning social media posts, videos, cards, flyers, photo collages & more. No design experience or expertise? No problem! From photo editor to collage maker, to logo maker - we made Canva really simple & easy for everyone. STUNNING TEMPLATES</div><div><br></div><div>4. Notion: Notion is a workspace that adapts to your needs. It's as minimal or as powerful as you need it to be. Notion continues to be the easiest way to get information centralized somewhere and shout it out to someone else. For us, that's extremely important because half our team is remote. Notion's ease of use is one of its hallmarks.</div><div><br></div><div>5. Bubble:Bubble is the most powerful no-code platform for creating digital products. Build better and faster. Demonstrate your idea before making an investment in technical resources. Build customer-facing web platforms and internal tools in hours instead of months. Empower anyone on your team to make changes, not just developers. </div><div><br></div><div>6. CopyAI: Content Marketer & Blogger. Copy.ai has more specialized tools than any other AI copy tools. The outputs (copy) generated are higher quality and more intuitive. The voice of the copy is more accurate and conversational.</div><div><br></div><div>7. Webflow: Webflow writes clean, semantic, standards-compliant code for you. Push projects further by easily integrating custom code or leveraging our CMS API. Creative power that goes way beyond templates. The Webflow Designer lets you build any website you can imagine with the full power of HTML, CSS, and Javascript in a visual canvas.</div><div><br></div><div>8. Gumroad: Gumroad is a self-publishing digital marketplace platform to sell digital services such as books, memberships, courses and other digital services. </div><div><br></div><div>9. Grammarly: Grammarly is a cloud-based typing assistant that reviews spelling, grammar, punctuation, clarity, engagement, and delivery mistakes. Millions trust Grammarly's free writing app to make their online writing clear and effective. </div>Unknownnoreply@blogger.com0tag:blogger.com,1999:blog-6698881918084302179.post-52522499822004793722022-02-18T23:34:00.000-08:002022-08-12T13:07:51.192-07:00Ten no-code tools every startup must know<p>1. GLIDEAPPS https://t.co/gVcvskgNaP</p><p><br></p><p>Create powerful apps and websites using basic spreadsheet skills.</p><p>Pricing: 500 data rows for free https://t.co/1mc2dRcYLy</p><p><br></p><p>2. WEBFLOW https://t.co/3Cq1MiyXaj</p><p>Create professional, custom websites in a completely visual canvas with no code.</p><p>Pricing: Free + Paid https://t.co/NlJ9wBwTeA</p><p><br></p><p>3. NOTION https://t.co/XVYX1wpZEB</p><p>Notion is a workspace that adapts to your needs. Write, plan & get organized in one place. </p><p>Pricing: Free + Paid https://t.co/WTv8zV9JGf</p><p>4. CANVA https://t.co/ium3zYI5d5</p><p>Create beautiful designs with your team. Design, share and print business cards, logos, presentations and more.</p><p>Pricing: 250,000+ templates for free https://t.co/vqDmLuuDd9</p><p><br></p><p>5. ZAPIER https://t.co/qMxUVizT5Q</p><p><br></p><p>Easy automation for busy people. Moves info between your web apps automatically, so you can focus on your most important work.</p><p><br></p><p>Pricing: 100 tasks /mo for free https://t.co/kZlHGoo07m</p><p><br></p><p>6.BUBBLE https://t.co/4yv9hYFJDc</p><p><br></p><p>Bubble is the most powerful no-code platform for creating digital products. Build better and faster.</p><p><br></p><p>Pricing: Free + Paid https://t.co/0JuOexG2v</p><p><br></p><p>7. CARRD https://t.co/By5HnH6Eb7</p><p><br></p><p>Build one-page sites for pretty much anything. Whether it's a personal profile, a landing page to capture emails, or something a bit more elaborate.</p><p><br></p><p>Pricing: Free + Paid https://t.co/sdR7GQNEoQ</p><p><br></p>Unknownnoreply@blogger.com0tag:blogger.com,1999:blog-6698881918084302179.post-43415063925952981332021-09-16T05:39:00.002-07:002021-09-16T05:39:41.897-07:00KRA launches Tims system<p> </p><div class="separator" style="clear: both; text-align: center;"><iframe allowfullscreen='allowfullscreen' webkitallowfullscreen='webkitallowfullscreen' mozallowfullscreen='mozallowfullscreen' width='320' height='266' src='https://www.blogger.com/video.g?token=AD6v5dzLwcRXWJg1GyxuYebGlSG8btA2Hc0EFzl5e6ykGStuN95L7CXseP-5WH1IGGhg-VKDCGFvvAAgs1ejviyYxw' class='b-hbp-video b-uploaded' frameborder='0'></iframe></div><br /><p></p>Unknownnoreply@blogger.com0tag:blogger.com,1999:blog-6698881918084302179.post-31458326606676468292018-09-15T16:22:00.000-07:002018-09-15T16:22:00.767-07:00A day of reckoning <div dir="ltr" style="text-align: left;" trbidi="on">
Today, I helped my kid learn the art of signature. It was the first time that she signed on something. The day signifies the beginning of one's attestation to one's own decisions. A milestone in one's life.<br />
<br />
Later, on the same form, there was a need to mention an emergency contact. I was debating whether it should be me or someone unrelated, because for all practical purposes an emergency would find all of us together, so it makes sense to say put out a number of our family in India? Well, then just to close the matter, I proceeded to put my own name on it.<br />
<br />
It was then, that rather casually, my daughter made a statement. That her mom is more of a parent to her than I am. Because she attends more PTC than I do.<br />
<br />
I think there's nothing childish about this line of thought. This is what we induce to. Our children day and night Heck she's been on a small level of rebellious behaviour i guess, what with the never ending forced off from school, given my frequent transfers and other such things. Yesterday she talked back to her mom about how it wasn't for her that she had woken up all night but for some selfish purposes and I followed It up with a good lecture of how moms are the best friend to anyone. If we parents get into an argument, like everyone does, I silently ask my daughter to take her mother's side, because she needs the support more than I do. I am strong enough to manage, I tell her.<br />
<br />
Anyway so its perhaps quite right that she chose to verbally denounce my claim on her. It was a bit of a shock as it was pretty unprovoked... It was a rebel against patriarchy perhaps ? A rebel and a feminist! Quite a combination that would be.<br />
<br />
At another level, I do fell a bit liberated. I have no remaining emotional encumbrances in this life than to provide enough money for my family. Life has come full circle. Long decades back I knew I was far too fond of myself to love another person more than myself. It's a self fulling prophecy when others feel the same way!<br />
<br />
PS : I didn't react to anything. What would I say!? </div>
Unknownnoreply@blogger.com0tag:blogger.com,1999:blog-6698881918084302179.post-91035744499168715042016-11-27T14:19:00.001-08:002016-11-27T14:19:13.137-08:00War is coming<p dir="ltr">Something dark is swelling up in the world. A fear, a dangerous fear, boiling over into anger and repulse.</p>
<p dir="ltr">The kind of fear our forefathers would have seen before. The kind that leads down a road we know too well. No need to go too far back in time. Genocides of the Partition, Bangladeshi Hindus, Kashmiri Pandits are still fresh.</p>
<p dir="ltr">The signs are happening everywhere, all at once. People are rising to the clarion calls of protecting themselves in their homes. Brexit in the UK. Trump in the US. Far-right revival in Europe. Nationalism in Russia. Machoism in China and India. Sectarianism in the Middle East. People are reacting to fear, everywhere you look.</p>
<p dir="ltr">Where did this come from? Why are people everywhere turning to nationalism? Why does everyone suddenly want to draw a line between themselves and the rest of the world?<br>
Because of fear. A single, gripping fear that’s sweeping the world, from England to Ohio, from Vienna to Moscow to Raqqa to Delhi to Rome.</p>
<p dir="ltr">The fear that the multi-culturalist globalist establishment isn't really working in our favour as we were promised it was. That it doesn’t have space for all of us, and ceded that space in pandering to the Jihadist. Fear that we and our way of living has been sold out to the Muslims in the guise of liberalism.</p>
<p dir="ltr">This fear is very real, and it’s driving people to fight for their place in a suddenly shrinking future. What we’re seeing them do, all over the world, is decide who they’re willing to fight alongside, and who they’re not.<br>
What we’re watching is the world draw battle lines. And on one side is the maudering jehadi who has no religion, baying for your kafir blood.</p>
Unknownnoreply@blogger.com0tag:blogger.com,1999:blog-6698881918084302179.post-83351085511115641192016-09-23T13:53:00.001-07:002016-09-24T14:07:35.380-07:00
Growing Up in a Hostile Environment: Screwed up Priorities of some Governments <p dir="ltr">Welfare Vs Hostile States.</p>
<p dir="ltr">Here's the situation : Immigration status is not easy in some countries. I plod through the process even after 6 months of having being in a country, with a reputable organisation. I have stayed away from my family because I did not want my kids academic year to be wrecked, so their immigration applications could be done parallely in good time. Education has always been my top priority...</p>
<p dir="ltr">Well, so last 2 month of her academic break should have been enough for her transit to the new country, or so I calculated. She has a place in school. I was pretty sure i had every thing covered when she passed her test and secured an admission. In another few weeks she should get her permit too. But there was a surprise on day-1 of new term. The school refused to let her take classes,  because her immigration status is not right.  And no, they cannot allow any grace time even if I can prove that the permit has been applied for. They used to in previous years, like my predecessor had suggested, but not anymore. The state has it's priorities. They raid the schools regularly to catch children of incorrect immigration statuses benefitting from a highly costly private education, paid for by legal immigrant parents. Government cannot allow such transgressions! The fines on schools are prohibitive.... Compliance before education!          </p>
<p dir="ltr">There is more in terms of benign rules.... One cannot bring dependents without 6 months of salary proof in the country. You got that right...!!!  you have to be a forced bachelor for at least 6 months by law. Also, Your demonstrable financial capability isn't enough for the authorities, like it does in other logical countries. Your local salary account statement is the only record of interest.... The statement should show 6 months of salary being credited in your local salary account. Why, you ask? </p>
<p dir="ltr">Because you are assumed to be a blue collared expat worker who constitute the majority of the population in the country. (Corrollary assumption: Your salary is assumed to be your only subsistence. You are assumed to have no other money anywhere in the world. You are assumed to be in complete mercy and control of your employer etc). All laws are devised on that assumption. And no, you cannot bring child without (one of) your spouse/(s) in tow. (to be Fair, that's also in other countries, I guess).  For me it creates a problem, because all your passports should be valid for next 6-months. And the 6-month transit has already brought one of ours' close to expiry! We are staring at a short term settlement which is as uncertain as my bank account balance nowdays! 😂 You cannot change jobs without employer NOC. You cannot leave the country without employer approval(exit permit). The list is long.... the hostility is deep.....</p>
<p dir="ltr">Anyway, The situation of my child got me thinking a bit. Here's the thing. What kind of country makes young children suffer through no fault of their own? Aren't many having their lives marred, including those who have only ever known this country as their home? </p>
<p dir="ltr">There's much to trouble about, but I shall focus on the conditions of my child's case study presented above and restrict myself to the right of children and the treatment to it worldwide.</p>
<p dir="ltr">Most respectable humane countries of the world are signatory to the UN Convention on the Rights of the Child where the rights of children are respected, regardless of their immigration status. However,  there remain the rogue countries of slave-owning cultures where immigration policies have a damaging impact on vulnerable groups including and especially children.</p>
<p dir="ltr">Let's have a look at the policy of education of some countries where children of doubtful immigration status is concerned. </p>
<p dir="ltr">In United States, All children are entitled to equal access to a public elementary and secondary education, regardless of their or their parents' actual or perceived national origin, citizenship, or immigration status. http://www2.ed.gov/policy/rights/guid/unaccompanied-children.html</p>
<p dir="ltr">In Canada, the Immigration and Refugee Protection Act says Schools cannot refuse to admit children. It is against the law for a school to refuse to admit a child who is under 18 years of age only because the child or the child’s parent or guardian is in Canada without immigration status. http://www.cleo.on.ca/en/publications/rightschool</p>
<p dir="ltr">In the United Kingdom, When you deal with an application for a child who is not a UK national, You must not refuse a school place simply because of doubts about the child’s immigration status.https://www.gov.uk/guidance/schools-admissions-applications-from-overseas-children.</p>
<p dir="ltr">The European Union member states have all <u>ratified</u> The United Nations Convention on the Rights of the Child of 1989 makes primary and secondary education available to all children in a country. Children first and foremost: the “best interests of the child”, says the convention. </p>
<p dir="ltr">According to the Convention on the Rights of the Child, the “best interests” principle applies to all children without discrimination, including foreign children who are illegally residents: “all the rights laid down by the Convention on the Rights of the Child must be applied to all the children present in the State, including […] those who are there illegally”!  Obviously, this works for students awaiting permanent immigration status. http://briguglio.asgi.it/immigrazione-e-asilo/2002/dicembre/oss-stc-comun-rimpatrio.html</p>
<p dir="ltr">However, There are these countries which have draconian laws as a way of buying more immigrants to contribute their manual labour but dissuading their families from coming. Most immigrants of the labour class are forced into the hands of rogue landlords and exploitative relationships. This is the kind of immigration that is desired and coveted. And high turnover ensures that limbs are young and comfort is low. Hence laws that encourage people to leave early are in order. The clever HR policy of wanting an immigrant's body's toil, not him as a human being. More unproductive mouths to feed (or dependants) are the last thing an expat is expected to bring to an economic workplace (the country). </p>
<p dir="ltr">How a country, treats children is a good measure of the values of a society. Aspirations of these countries are sky high - hosting marquee world events and tall towers and Formula races. Are these aspirational countries not failing to live up to the high standards that they rightly would set for others like the European Union, which has accepted millions of refugees from their troubled neighbourhood? (By the way, how many accepted by the n'hood itself?)</p>
<p dir="ltr">It is extremely concerning that rather than trying to improve the ways in which to support children, the governments instead make their countries a more hostile place for them to grow up in.</p>
<p dir="ltr">Disclaimer : The purpose of this article is not to criticise any country or law, neither do I have the competence to comment on legal matters, but to take a decision about whether the environment actually suits the self. Let me know your views. </p>
<div class="separator" style="clear: both; text-align: center;"> <a href="https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEjMD-AdHjJE7ttAR17BbOad09bY1OZPwyqAvjU5NZSEM9l8xmQ2ZPdrUvsoOLOe4yJfzVsH6w30_EESEWmDqlok5qQfsXJD9s0jQ22xqdcYXz60bpULXCOKhunYMp6IF4eHfxqEbBMJPZ0/s1600/20160922_142233.jpg" imageanchor="1" style="margin-left: 1em; margin-right: 1em;"> <img border="0" src="https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEjMD-AdHjJE7ttAR17BbOad09bY1OZPwyqAvjU5NZSEM9l8xmQ2ZPdrUvsoOLOe4yJfzVsH6w30_EESEWmDqlok5qQfsXJD9s0jQ22xqdcYXz60bpULXCOKhunYMp6IF4eHfxqEbBMJPZ0/s640/20160922_142233.jpg"> </a> </div><div class="separator" style="clear: both; text-align: center;"> <a href="https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEindaIlGFAIOTbRU8-dg3zdmehCNRHHfORr7b-RK1V2VYu0AQWrN5n4CHv1HMhnHLzt-_tEKL3gExsrNOSSm_1mG1uIdySy8Hgrhu-RND2b9_mhf2hVZEE4p8hjHtRFqaMiJyQCmcOYOqc/s1600/Screenshot_2016-09-25-00-05-47.png" imageanchor="1" style="margin-left: 1em; margin-right: 1em;"> <img border="0" src="https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEindaIlGFAIOTbRU8-dg3zdmehCNRHHfORr7b-RK1V2VYu0AQWrN5n4CHv1HMhnHLzt-_tEKL3gExsrNOSSm_1mG1uIdySy8Hgrhu-RND2b9_mhf2hVZEE4p8hjHtRFqaMiJyQCmcOYOqc/s640/Screenshot_2016-09-25-00-05-47.png"> </a> </div>Unknownnoreply@blogger.com0tag:blogger.com,1999:blog-6698881918084302179.post-89079559992313626992016-09-08T15:50:00.001-07:002016-09-08T15:50:14.685-07:00Life in the deserts. Or the lack of it. <p dir="ltr">So here's the thing.</p>
<p dir="ltr">It doesn't matter who you are or what you earn, One can't bring one's family to the country before proving 6 months of gainful employment. </p>
<p dir="ltr">Majority of the so-called expats, which is about 85% of the population, can't even afford the threshold, so the whole place is swarming with frustrated, overworked, underpaid workmen with severely restricted lives. Men only. Because women and children await the expat remittances our home country is so proud of. Oldies aren't allowed in either. Families are discouraged. Health is screened before immigration. Only fit males who can labour well. It's a carefully constructed "perfect world". Hitler would die to learn, if he weren't already. A lot of people would learn too, had they survived the harsh conditions. But they constructed the edifice and facades.... the exteriors hide the grim reality of empty realty. The interiors hide the unforgiving deserts outside. </p>
<p dir="ltr">इस दार-ए-तरक्की के अंदाज निराले हैं।<br>
रास्तो पर उजाले हैं, जहनों में अंधेरे हैं।।</p>
<p dir="ltr">To top it all, 6 month moratorium extends to everything about the family. Even school. Surprise raids will catch the erring schools imparting that overpriced education to illegal families who managed to sneak in with their husbands. Such is the rules. </p>
<p dir="ltr">Sun is clouded today.  Just another sandstorm. The dust clouding visibility. Keep working. Put your head down, bury it deep.</p>
<p dir="ltr">Anybody wants to join this fabulous wonderland! Time of my life! </p>
Unknownnoreply@blogger.com0tag:blogger.com,1999:blog-6698881918084302179.post-75599934250326226482016-07-21T06:12:00.000-07:002016-09-08T15:50:39.775-07:00Prasun Chaturvedi Interview<div dir="ltr" style="text-align: left;" trbidi="on">
Interview published in TCS magazine.<br />
<br />
<div class="separator" style="clear: both; text-align: center;">
<a href="https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEgk9ZjejUAOQ2lOpbxzhYmt7D9eTpm0e7pbKeIoUzDibWi7aiDNCBlH-rrdDoxJnGH-THMP5K4Sp-hCTjpxzTFev6DRx9DkBqukHBRXgYPEMvS1akYRjm3spzK9nxvATpe4VPJ2rEbNwyw/s1600/Prasun+-+1.png" imageanchor="1" style="margin-left: 1em; margin-right: 1em;"><img border="0" height="640" src="https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEgk9ZjejUAOQ2lOpbxzhYmt7D9eTpm0e7pbKeIoUzDibWi7aiDNCBlH-rrdDoxJnGH-THMP5K4Sp-hCTjpxzTFev6DRx9DkBqukHBRXgYPEMvS1akYRjm3spzK9nxvATpe4VPJ2rEbNwyw/s640/Prasun+-+1.png" width="496" /></a></div>
<br />
<div class="separator" style="clear: both; text-align: center;">
<a href="https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEjwqJLqSYzYuXL1ZZ36cem6YDvRPEr5dkUh2n-icQs-O0wo1K5UzrppNDJ33yQzmCH4EwI5fZRqGuIT-JC-QD_lYryUcSlch5344AhdlWSeMXfX9gZ-oawbt-XQ__CHp-xf_WuXSaYMf6Y/s1600/Prasun+-+2.png" imageanchor="1" style="margin-left: 1em; margin-right: 1em;"><img border="0" height="640" src="https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEjwqJLqSYzYuXL1ZZ36cem6YDvRPEr5dkUh2n-icQs-O0wo1K5UzrppNDJ33yQzmCH4EwI5fZRqGuIT-JC-QD_lYryUcSlch5344AhdlWSeMXfX9gZ-oawbt-XQ__CHp-xf_WuXSaYMf6Y/s640/Prasun+-+2.png" width="494" /></a></div>
<br /></div>
Unknownnoreply@blogger.comtag:blogger.com,1999:blog-6698881918084302179.post-35550163888267869932016-03-01T09:14:00.001-08:002016-03-13T01:55:47.618-08:00Modi has changed gears - Take no prisoners.<p dir="ltr">http://myvoice.opindia.com/2016/03/modi-seems-to-have-changed-gears-politically/</p>
<p dir="ltr">The sudden spat of exposes of Ishrat Jahan witnesses, the ED finding all of PC son's illgotten wealth right during the Parliament session just shows Modi was accumulating these facts do4 a long time to strike at the right time. </p>
<p dir="ltr">Parliament wasn't allowed to function again as expected.  It is Rahul's brainwave to hold the government to ransom and not allow it to do any constructive legislative work , hence, in his mind discrediting the government. Especially since expectations are high from Modi and people are getting restive. Acched Din, Rahul thinks, can be beaten by Jumla sarkar.</p>
<p dir="ltr">Modi has given a teaser by accountability actions against PC... a direct warning to Sonia that noone is beyond reach. This damocles sword is the last attempt at getting some much-needed cooperation from government. Last attempt, because the by polls to Rajya Sabha due in July, given the current state mathematics, is expected to push Congress down to less than 60 seats. This should be enough for Government to conduct business with the help of friendly parties like AiADMK, BJD or even SP.</p>
<p dir="ltr">Given the threat hanging on PC, we can expect parliament to start cooperating a bit from now on. Financial bills will probably go through. The very top legislation business may slide through. Obviously the optics if confrontation will continue, but with the right prodding at the right time at the right places, government may succeed to buy temporary peace. I have suspicion that GST is just a tool to discredit the opposition. They quickly fell to Modi's trap in vociferously opposing false flags like Land Acquisition and GST, not half as necessary as it is made out to be. If the opposition now support GST, they'll have to eat humble pie. If not, modi will play victim, and further turn the screws. </p>
<p dir="ltr">Modi has also got JNU as his 2nd Godhra moment. Deep polarisation along nationalistic lines has been achieved. This is Vintage chess minds at play here (Amit Shah has long been an avid chess player). It is evident that govt is changing gears. "Take no prisoners" is the slogan but getting opposition to surrender may be on the agenda. The situation is Good for the economic agenda in my opinion.  </p>
<p dir="ltr">If the economy succeeds in creating the new rural demand as it's engine, as budget 2016 focuses, Sonia and her Suit-Boot opposition will be history. Sonia isn't going to take it lightly either. After a temporary ceasefire to buy reprieve from government action, More identity based politics may be in the offing to break the nationalistic narrarive. Will it create enough disturbance in country to trip the economy and hence Modi? Will Modi use his brahmastra of action agianst his arch rival? Will it be too late to needle the Queen? Interesting times ahead! </p>
Unknownnoreply@blogger.com0tag:blogger.com,1999:blog-6698881918084302179.post-35723169093282203312016-02-24T09:31:00.000-08:002016-02-24T09:31:03.147-08:00Pampore Terrorist Attack - Part III<div dir="ltr" style="text-align: left;" trbidi="on">
- via Whatsapp<br />
<br />
PAMPORE TERRORIST ATTACK - A Special Kind of Failure<br />
Manvendra Singh<br />
<br />
<br />
“`I don't think they play at all fairly,' Alice began,' ...and they don't seem to have any rules in particular; at least, if there are, nobody attends to them.“ Alice, in this case, is the one lost in Wonderland. And she has an important lesson when it comes to the recently concluded, and deeply tragic, operation in Pampore, Jammu and Kashmir.<br />
All the top-notch military précis, training manuals and doctrines don't prepare a conventional army leadership when it comes to employing Special Forces. Those lines from Alice encapsulate all that there is to know about how the Special Forces play -and how to play with them.<br />
<br />
Special Forces are essentially anathema to a regular conventional professional army: a small number of highly trained and motivated lads, doing operations disproportionate to their numerical strength. An army is about mass; the Special Forces about less being more. The twain don't meet.<br />
<br />
But an army has to tolerate them, because they exist. And they exist because there are individuals crazy enough to feel comfortable and confident with fewer around in taking on more than what they should. This, in essence, is the basic difference between a conventional army , and Special Forces.<br />
<br />
India is not unique in this asymmetry . All professional (or even ceremonial) armies have faced this dilemma and `dispute'. Forward-thinkers will find a way out by accepting the reality that the `crazies' of any society can be an asset and should be used as such.<br />
<br />
They leave egos aside, join the future, and give the Special Forces their space to operate. Some armies, especially the ones encumbered by their large bureaucratically driven system, deny the future and this logic. And then they pay a heavy price.<br />
<br />
Pampore extracted that far too heavy price. In the public domain, it will remain an operation in which three terrorists were holed up in a building, killed three Special Forces troopers who tried to neutralise them. The public domain will look at, and remember, the operation as even-steven.<br />
<br />
But it wasn't so, as the army was the loser. And it wasn't so simply because the operation was directed by those who didn't know how to employ Special Forces, and despite not knowing, insisted on telling the three Specials who died and their comrades who participated in the operation, how to do it.<br />
<br />
Alice had figured it out perfectly . If you want them to play , let them play as they want to, fairly or otherwise. They have their rules, even if the rule book is indecipherable by most people. Never tell them this is the playground and play according to these rules. Because they can't. If they could, they'd be wearing different regimental insignia.<br />
<br />
The insignia Special Forces wear on the right pocket compels them to think, feel, breathe and play differently .The commanders of war generally , and those in Pampore specifically , believe otherwise, and three precious lives were lost. Disproportionate to the three terrorists neutralised.<br />
<br />
The terrorists holed up in the government building had their military psychology lessons correct. They released the civilians to continue with the charade of fighting for them. They pretended they had vulnerable hostages, and compelled the commanders to order an assault.<br />
<br />
In a hostage situation, time is essential as delay means deaths. But in Pampore, an assault was ordered before it was clear whether there were hostages, where they were, and how many terrorists held them. Because Pampore is Srinagar and a multistorey government building meant the exigency was getting the hostages out fast.<br />
<br />
Which is the first, fatal, mistake when it comes to military planning, special operations in particular. Never have a time-frame for concluding an operation, even if anti-national sloganeering is being aired by TV channels who don't think much of it.<br />
<br />
Authority is about having the spine and the hide to take the barbs without losing the larger picture. In Pampore, alas, that was not the case. The TRP game came to play in a life-and-death situation, when there was no need.<br />
<br />
Left to the Special Forces, both outstanding units, they'd have dealt with the operation differently . Room entry is part of their training. But multiple rooms and multistoreys are a different game altogether. Special Forces have the ability to adapt to a new challenge for which they haven't trained.<br />
<br />
So, let them adapt, find the route, even if it is not a `fair' one. At least precious lives wouldn't have been lost, for the sake of time, which in the end doesn't tell. Only statistics do. And in that the army botched up. Badly .The writer is editor , Defence and Security Alert. Views are personal</div>
Unknownnoreply@blogger.com0