| Weekly Schedule and Class Notes |
- Lecture notes : all notes will be posted in this section.
- It is your responsibility to download, print and bring the notes to the class. Notes will be available 24 hours before each class.
|
Week |
Date |
Topic |
Reading |
Assignments |
Notices and Dues |
Notes |
|
1 |
03/04 |
1. Introduction |
Lecture Note |
Homework #1 |
Due for Homework #1 [9:00PM, March 10th] |
|
|
2 |
03/11 |
2. Reinforcement Learning 3. Value iteration, policy iteration, policy search 4. Value Vs Q 5. Q and Policy update 6. Temporal distance learning |
Lecture Note |
Homework #2 (model collapse)
|
|
|
|
3 |
03/18 |
7. Model collpase 8. Advantage & Asyncronous 9. A3C, AC, A2C |
Lecture Note |
|||
|
4 |
03/25 |
11. AC, deep understanding 12. On-Policy Vs Off-Policy |
Lecture Note |
Homework #3 (New types of DRL) |
||
|
5 |
04/01 |
13. Student Presentation (II) 14. Off policy & Importance sampling |
Lecture Note |
|||
|
6 |
04/08 |
15. Type of DRL 16. Value based mechanisms: DQN, DDQN, Dueling, Rainbow DQN 17. Poicy based mechanisms : AC,A2C, A3C, TRPO, PPO |
Lecture Note |
Homework #4 (DRL architecture design) |
||
|
7 |
04/15 |
18. AC-based Hybrid mechanism: DDPG, TD3 19. Soft AC architecture 20. Stochastic Process & OU process |
Lecture Note |
|||
|
8 |
04/22 |
Midterm exam (I) |
|
|||
| 9 |
04/29 |
Midterm exam (II) |
Lecture Note |
|||
|
10 |
05/13 |
Reading day |
|
|||
|
11 |
05/20 |
22. Kolmogorov-Arnold Network (KAN) 23. Kolmogrov-Arnold Representation Theorem 24. Bezier nonlinear fitting 25. B-spline based nonlinear fitting (I) |
Lecture Note |
|||
|
12 |
05/27 |
26. B-spline based nonlinear fitting (II) 27. NUBS, NURBS 28. KAN (II) |
Lecture Note |
|||
|
13 |
05/21 |
29. KAN (III) |
Lecture Note |
|||
|
14 |
06/03 |
Reading days |
|
|||
|
15 |
06/10 |
30.Summary of Deep Reinforcement Learning 31. Stochastic Processes - Brownian Motion, Ito Integral |
Lecture Note |
|||
|
16 |
06/17 |
32. Stochastic Processes and DRL |
Lecture Note |
|||
| 17 |
06/24 |
Final exam |
|