Lectures
Unit 1: Foundations
- L1
-
- Overview: What is ML?, real-life applications, linear and logistic regression.
- Slides TBA
- L2
-
- How ML Models Learn: Supervised learning, loss functions, gradient descent, classifiers.
- Slides TBA
- L3
-
- Building a Neural Network: Neural networks, backpropagation, activation functions.
- Slides TBA
- L4
-
- Making Neural Networks Learn: Optimization, dropout, initialization, regularization.
- Slides TBA
Unit 2: Vision and Sequences
- L5
-
- Computer Vision: Convolutional neural networks, pooling, feature maps.
- Slides TBA
- L6
-
- Sequential Models: ResNets, recurrent neural networks, long short-term memory networks.
- Slides TBA
Unit 3: Attention and Transformers
- L7
-
- Attention: Attention, queries, keys, and values, self-attention, cross-attention, tokenization.
- Slides TBA
- L8
-
- Transformer Architecture: Multi-head attention, positional encoding, the Transformer.
- Slides TBA
Unit 4: Generative and Modern AI
- L9
-
- ML Without Labels: Unsupervised learning, K-means clustering, pre-training, self-supervised learning.
- Slides TBA
- L10
-
- Generative Models: Generative modeling, GPTs, variational autoencoders, Vision Transformers.
- Slides TBA
- L11
-
- Modern AI Systems: RLHF, prompt engineering, ReAct, chain-of-thought reasoning.
- Slides TBA
Unit 5: Reinforcement Learning
- L12
-
- Reinforcement Learning: Agents, Markov decision processes, value functions, Q-learning.
- Slides TBA