Primary Textbook: Artificial Intelligence: A Modern Approach by Peter Norvig and Stuart J. Russell
Module 1: Search and Planning
Module 2: Learning
| 9 |
Feb 4 |
Feb 3 |
Introduction to Machine Learning |
|
M01V01–V04: What is ML?, Types of ML, Classification vs Regression |
| 10 |
Feb 9 |
Feb 9 |
Data Foundation (HTML | PDF) |
ML paradigms, features/labels, train/test split, sklearn |
M01V03: Train vs Test |
| 11 |
Feb 11 |
Feb 10 |
Supervised Learning (HTML | PDF) |
Linear regression |
M05V01–V07: What is LR?, Matrix Form, Error & Objective, Normal Equation, Worked Example, Outliers, Basis Expansion |
| 12 |
Feb 16 |
Feb 16 |
Supervised Learning (HTML | PDF) |
Linear regression |
(same as above) |
| 13 |
Feb 18 |
Feb 17 |
Supervised Learning (HTML | PDF) |
Logistic regression |
M07V01–V08: Why LR Fails, Sigmoid, LR Model, Decision Boundary, Why Not MSE?, Cross-Entropy, Gradient & Learning, Non-Linear Boundaries |
| 14 |
Mar 9 |
Mar 9 |
Neural Networks — MLP (PDF), Notebooks: NN Power, Perceptron |
Perceptrons, logic gates, XOR, MLP, activations, forward propagation |
M08V01–V09: Features to Learning, Perceptron, Logic Gates, XOR Problem, Activations, MLP, Forward Prop, Vectorized, XOR Solved |
| 15 |
|
Mar 10 |
Neural Networks — Autograd (PDF), Notebooks: Autodiff, FC Visualizer, Autograd Playground |
Backpropagation, autograd, training loop |
M09V01–V05: Why Autodiff?, Comp Graphs, Backprop, Full Backward Pass, Autograd Engine |
| 16 |
Mar 16 |
Mar 16 |
Next Token Prediction (PDF), Notebook: Next Char, Demo: GPT Demo |
Embeddings, next character prediction, MLP for generation |
M10V01–V08: What is NTP?, Indian Names Dataset, Training Set, Char Embeddings, Embedding Lookup, Full Architecture, Training, Generating Names |