Pragnya Yenuga’s Post

🚀 Day 1 of Learning Transformer Architecture day1:self attention Today, I took my first step into the fascinating world of Transformers, one of the most powerful models in deep learning. I will try to share my notes soon. I started by understanding Self-Attention, a core concept that enables Transformers to understand the relationships between different parts of a sequence, regardless of how far apart they are. It was intriguing to see how it calculates the importance of each token in a sentence to every other token. I find this small video helpful to understand linear relationships in the transformer’s positional encoding: https://lnkd.in/gWDAzX-w #Transformers #DeepLearning #SelfAttention #MachineLearning #NLP #LearningJourney #AI

🚀 Ever wondered how AI works behind the scenes? 🤖 Curious about Crew AI and the magic of Retrieval-Augmented Generation (RAG)? 🔍 Check out this video where you can quickly dive into how AI RAG systems process vast amounts of data and make decisions in real-time! 🧠💡 From understanding language to retrieving relevant info on the fly, this video is a backstage pass to the AI powerhouse that makes it all happen. 🎥 Whether you’re a tech enthusiast or just curious about how these systems work, I hope this one’s for you! 👨💻👩💻 💼 #AI #MachineLearning #CrewAI #RAG #Tech #Innovation #ArtificialIntelligence #DeepLearning #DataScience #NLP #SoftwareDevelopment

"Transforming Data with Feature Engineering: A Key to Model Success" 🔍 Feature Engineering: The Secret Sauce to Enhancing Your Models! Feature engineering is crucial for improving model performance. Here’s how you can leverage it: Create New Features: Derive features from existing ones (e.g., creating "age" from "birthdate"). Feature Selection: Choose the most relevant features to reduce noise (e.g., using correlation matrices or feature importance scores). Normalization & Scaling: Ensure your features are on a similar scale (e.g., Min-Max Scaling or Standardization). 💡 Quick Tip: Experiment with different feature transformations and selections to see what works best for your model. #AI #ML #DeepLearning #NLP #GenerativeAI #DataScience #MachineLearning #FeatureEngineering

"Boosting Model Performance: A Quick Dive into Hyperparameter Tuning" 🚀 Tip for AI/ML Engineers: Maximize Model Performance with Hyperparameter Tuning! Hyperparameter tuning can make or break your model's performance. Here’s a quick guide to get you started: Understand Hyperparameters: These are not learned from the data but set before training (e.g., learning rate, tree depth, etc.). Popular Techniques: Grid Search: Exhaustive search over specified parameter values. Random Search: Random combinations of parameters for faster results. Bayesian Optimization: Smarter search by balancing exploration and exploitation. 🎯 Pro Tip: Start with Random Search for quicker insights and then fine-tune with Grid Search. #AI #ML #DeepLearning #NLP #GenerativeAI #DataScience #MachineLearning #TechTips

Anchoring bias describes the common human tendency to rely too heavily on the first piece of information offered (the "anchor") when making decisions. In decision-making, anchoring occurs when individuals use an initial piece of information to make subsequent judgments. In AI, anchoring bias can happen in various ways, primarily through the data used to train machine learning models and the initial parameters set by the developers. When setting up AI models, the initial parameters and assumptions can act as anchors. Example: In a recommendation algorithm, if the initial algorithm overemphasizes certain user behaviors, it might create a feedback loop where the user is only recommended similar items, reinforcing initial consumption patterns. #DataScientist #MachineLearning #NLP #LLM #DeepLearning #DataScience #Statistics #AI #LanguageAI #GenerativeAI #AIforHealth #ResponsibleAI #BiomedicalResearch #BiasInAI

𝐑𝐞𝐚𝐝𝐢𝐧𝐠 𝟓𝟎+ 𝐑𝐞𝐬𝐞𝐚𝐫𝐜𝐡 𝐏𝐚𝐩𝐞𝐫𝐬 𝐨𝐫 𝐂𝐨𝐦𝐦𝐮𝐧𝐢𝐜𝐚𝐭𝐢𝐧𝐠 𝐰𝐢𝐭𝐡 𝟓𝟎+ 𝐑𝐞𝐬𝐞𝐚𝐫𝐜𝐡 𝐏𝐚𝐩𝐞𝐫𝐬: 𝐈𝐬𝐧'𝐭 𝐭𝐡𝐞 𝐂𝐡𝐨𝐢𝐜𝐞 𝐎𝐛𝐯𝐢𝐨𝐮𝐬? Tired of sifting through countless reports, literature reviews, etc, trying to extract valuable insights and information? What if I told you there's a better way? AI tools have revolutionized the way we interact with research papers, allowing users to have Q&A sessions with their collected research, streamlining the process! Here are my top 5 favourite AI tools that do this: 1. SCISPACE 2. ChatPDF 3. Elicit 4. docAnalyzer.ai 5. OpenRead With these AI tools, you can make your research process more efficient and effective! 🌟Have you tried any of these yet? Start today and discover a new way to optimise your research process!🌟 Follow for more great tips 💡🎨✍️!!! #AIinResearch #ResearchTools #AcademicProductivity #ResearchPapers #NLP #MachineLearning #ResearchInnovation #ScienceCommunication

🔥How Generative AI sparks creativity? ☑️Generative AI sparks creativity by leveraging complex algorithms to autonomously generate new, diverse, and often unexpected outputs. ☑️By learning patterns and structures from vast datasets, these algorithms can produce novel content, whether it's images, text, or other forms of creative expression. ☑️The ability to combine and remix existing ideas in unique ways opens up new avenues for innovation and inspiration, fostering a dynamic interplay between human input and machine-generated output. ☑️Through this collaboration, Generative AI becomes a powerful tool for expanding creative boundaries and unlocking fresh perspectives in various fields. →Swipe through and discover the boundless possibilities of AI-generated wonders #datascience #artificialintelligence #machinelearning #deeplearning #nlp #generatieveai

🧠 🤖 𝐊𝐧𝐨𝐰 𝐲𝐨𝐮𝐫 𝐌𝐚𝐜𝐡𝐢𝐧𝐞 𝐋𝐞𝐚𝐫𝐧𝐢𝐧𝐠 𝐟𝐫𝐨𝐦 𝐲𝐨𝐮𝐫 𝐃𝐞𝐞𝐩 𝐋𝐞𝐚𝐫𝐧𝐢𝐧𝐠? If not, you’re not alone! Here are five of the most common AI terms defined… 𝑨𝑰: Systems called models that think and act like humans 𝑴𝑳: Models that learn from data to solve new problems 𝑵𝒆𝒖𝒓𝒂𝒍 𝑵𝒆𝒕𝒘𝒐𝒓𝒌𝒔: Digital brains that process data 𝑫𝒆𝒆𝒑 𝑳𝒆𝒂𝒓𝒏𝒊𝒏𝒈: Layering neural networks to solve complex problems 𝑵𝑳𝑷: Models that understand human language (think Siri or ChatGPT) I probably enjoyed distilling these terms more than I should. Reminds me of an old programming challenge called code golf, the aim of that being to solve a problem with the least amount of code possible. Glossary golf? I’m off to make another coffee… Have a different take on these? Share your thoughts below 😁👇 #AI #MachineLearning #DeepLearning #TechInnovation #NLP #AI101 #FutureOfTech

🌐 The Power of Chain of Thought in Large Language Models A new study reveals some exciting insights about Large Language Models (LLMs) and Chain of Thought (CoT) reasoning. Quick Takeaways: 👉 Length Matters: Giving LLMs more steps to think things through makes them smarter. 👉 Simplicity vs. Complexity: For simple problems, keep it short. For complex ones, longer reasoning helps more. 👉 Unexpected Findings: Sometimes, even long but not perfectly correct reasoning can be helpful. This research tells us how to make the most of AI's thinking process. 🔗 Check the paper: https://lnkd.in/dFVMUrXD #ai #machinelearning #techinsights #datascience #llm #ailearning #airesearch #NLP #LLM #opensource

AI for Everyone on LinkedIn: Whether you admit it or not, we have to rely on AI systems for our day-to-day activities, whether in #business, #education, #healthcare, or any other domain. I present you a brief history+timeline and perspective about #ai #machinelearning #deeplearning #computervision #nlp #chatgpt for everyone out there in layman's terms. The following topics are covered: ✡ Ket Terms: AI / ML / Data Science / Deep Learning / LLM ✡ How to develop intelligent systems? ✡ A Brief History and Timeline of AI ✡ Limitations of AI ✡ Future Directions #happylearning