"Artificial Intelligence, Machine Learning, Deep Learning" - Do you find these terms thrown around randomly and want to get started but don't exactly know how? Worry not! We bring to you DL Marathon - A full run of Deep Learning Madness, where we deep dive into various subtopics of deep learning like computer vision, natural language processing, reinforcement learning, etc. For each of these topics, we will pick a few relevant problems and help you understand them in detail. Further, we will be implementing each model during a live code-along to help strengthen your understanding!
Session Overview
- 14th Feb: Introduction - Basic fundamentals of ML (linear regression, gradient descent, backpropagation, etc), Introduction to Pytorch
- 16th-17th Feb: Computer Vision - Introduction to CNNs, Implementing simple Image Classification and Segmentation models, Visualization to build an intuitive understanding of how CNN's work.
- 18th Feb: Generative Adversarial Networks - Introduction to GANs (why and how?), implementing a simple handwritten digit Generation model.
- 22nd Feb: Reinforcement Learning - Introduction to RL, Q Learning for snake game, Deep Q Learning to play Visdom.
- 2nd - 3rd March: Natural Language Processing - How to represent words (TF-IDF, Word2Vec), Sentence Classification models, Sequential models (RNN, LSTM), Language Translation (seq-to-seq).
Pre-requisites
There are NONE at all. However, do expect a strong learning curve as we are trying to cover a lot of things at a decent pace. Don't worry we are starting right from the beginning.
Direct links to open in Google Collab can be found in the respective Notebooks
| Session | Recording | Content |
|---|---|---|
| 1 | link | ./intro/ |
All the session content and recordings will be shared in this repository. We strongly urge everyone to post their queries in the discussions tab rather than personally contacting us to avoid repetitions. This also helps build a large compilation of questions that would be useful for everyone to refer to at any point in time.