"ALL THINGS ML... Literally... Oversimplified" ~ A challenge that we have taken upon us and are extending it to you !

DISCLAIMER: 3 words, Useful - Fun - AI. We would like to stress on the useful part. We are extremely cautious about our sessions not ending up being something you can read in a blog in 5 minutes but you would have to sit through an hour in the event for the same. Exhausting, we relate ☹️

This session is to encourage everyone to start asking the right questions and cut through the hype and the black box approaches when it comes to -

⭐ ML Research vs Applied ML: You NEED to know the difference
⭐ Got a problem?: Make it a problem ML can solve
⭐ Critical Thinking: Got a problem ML can solve? Great! Now how to solve it. HOW TO ALIGN YOUR OBJECTIVES WITH YOUR LOSS FUNCTIONS?
⭐ Data Engineering: Sources, Formats, Models… How to piece it all together in the CORRECT CONTEXT?
⭐ Training Data: Where does it come from? Kaggle? Well yes.. But No. A discussion on the critical thinking and techniques required to create training datasets and avoid disasters taking real-life examples of Language Models & Self-Driving Cars.
⭐ Data Labeling: A sneak-peek with jargon you do need to be aware of
⭐ Dealing with Class Imbalance: Data-level methods and algorithm-level methods
⭐ Hands-On MLOps with PyCaret, MLFlow & DagsHub: Covers everything we learnt so far and also includes the rest - Data & Model versioning, Experiment tracking, Feature Engineering, Model Selection, Model Training, Model Deployment