Syllabus (Tentative)

Introduction:

Machine Learning Basics

1. Empirical loss minimization framework

2. Linear Classifiers

3. Nearest Neighbor Classifiers

4. Regularization

5. Non-linear Classifiers

6. Bias Variance Trade-off

7. Controlling classifier complexity

8. Digit recognition case study

Neural Networks

1. Back-propagation

2. Designing Neural Networks for Images

3. Training a Neural Network

4. Neural Networks for Detection and Dense Prediction

Image Generation

1. Variational Auto-encoders

2. Diffusion Models

3. Generative Adversarial Networks

Advanced Topics

1. Self-supervision

2. Neural Radiance Fields

3. Large-language Models