| Week | Date | Content | Exercise |
|---|---|---|---|
| 1 | 9/1 | Overview of the course | |
| 2 | 9/8 | 1. Statistical learning 2. Important probability and statistics concepts |
|
| 3 | 9/15 | Bayesian network and d-separation | Exercise 1 (due: 9/28 23:59:59) |
| 4 | 9/22 | 1. Naive Bayesian 2. Bayesian network modeling examples |
|
| 5 | 9/29 | Teachers' Day, NO CLASS | |
| 6 | 10/6 | Mid-autumn festival, NO CLASS | |
| 7 | 10/13 | 1. Markov Random Fields 2. Point estimation: MLE and MAP |
Exercise 2 (due: 10/26 23:59:59) |
| 8 | 10/20 | Exponential family and conjugate prior | |
| 9 | 10/27 | 1. Hypothesis testing 2. Midterm review |
|
| 10 | 11/3 | Midterm exam | |
| 11 | 11/10 | EM Algorithm: applications to GMMs (classic generative model) | Exercise 3 (due: 11/23 23:59:59) |
| 12 | 11/17 | Invited talk (TBA) | |
| 13 | 11/24 | 1. Monte Carlo methods 2. Markov Chain & MCMC (Metropolis-Hastings, Gibbs sampling) |
Exercise 4 (due: 12/7 23:59:59) |
| 14 | 12/1 | 1. Foundations of Reinforcement learning: Markov decision process (MDP) 2. Dynamic programming: policy and value iteration |
|
| 15 | 12/8 | 1. Deep reinforcement learning: value-based methods: deep Q-networks (DQN) 2. Policy-based methods: policy gradient and proximal policy optimization (PPO) |
|
| 16 | 12/15 | Final exam |