SP16: REINFORCEMENT LEARNING FOR AI: 32711

Home page

CS B659: Reinforcement Learning

News:

 

Basic info:

  • AI: Mrinmoy Maity
    • office: Lindley Hall 406
    • office hours: Tuesday 11am - 1pm
    • email: mmaity@umail.iu.edu

 

Class schedule

This schedule is tentative and may change throughout the semester.

Readings and assignment questions are from the main course text: Sutton and Barto Links to an external site.

Supplementary material can be found in Szepesvari's book Links to an external site..

These slides will be released progressively; look at the date on the first page to see how recently they have been updated.

Week Date Lecture Readings and Deadlines
1 Jan 12 Chapter 1: Introduction

 

-read Chapters 1 of Sutton and Barto

Jan 14 Download Chapter 2: Evaluative feedback

  

2 Jan 19 Chapter 2Evaluative feedback continued  Download Chapter2_2white.pdf

Though questions about chapter 1&2 due Jan 18th at 11:59pm 

Assignment #1 released (bandit problems and MDPs)

Jan 21 Chapter 3: The reinforcement learning problem Download Chapter3white.pdf
3 Jan 26

Chapter 3 continued Download Chapter3white_v2.pdf

Thought Questions about Chapter 3 due
Jan 28

Chapter 3 completed. Start of Chapter 4 Dynamic programming  Download Chapter4white.pdf

 

4 Feb 2 Finish Chapter 4: Dynamic programming  Download Chapter4white_v2.pdf
Feb 4 Chapter 5: Monte Carlo methods Download Chapter5white.pdf

 

5 Feb 9

Chapter 5: Monte Carlo methods Download Chapter5white_v2.pdf

 

Assignment #1 due

Assignment #2 released (elementary solution methods)

 

Feb 11

Course Projects Download Projects_Review.pdf

 

6 Feb 16 Chapter 6: Temporal difference learning Download Chapter6white.pdf
Thought Questions about Chapter 4 & 5 due 
Feb 18 Chapter 6: Temporal difference learning
7 Feb 23 Chapter 6:Review lecture, Download review_Ch2thruCh6.pdf

 

Thought questions about Chapter 6 due

Feb 25

Chapter 7: Eligibility traces

 

Assignment #2 due

Assignment #3 released (TD methods and eligibility traces)

8 Mar 1

Chapter 7: Eligibility traces Download TD_lambdaCh7.pdf

Mar 3 Chapter 7: Eligibility traces Download TD_lambdaCh7_v2.pdf

Thought Questions about Chapter 7 due 

9 Mar 8 Chapter 8: Planning and learning Download Chapter8.pdf
Mar 10 Chapter 8: Planning and learning Download Chapter8_v2.pdf
Project proposals due
10 Mar 15 No classes: Spring break
Mar 17 No classes: Spring break

 

11 Mar 22 Chapter 9:On Policy Prediction with Approximation: Download Chapter9_white.pdf

Assignment #3 due

Assignment #4 released (tabular planning methods)

Mar 24 Chapter 9 On Policy Prediction with Approximation:  Download Chapter9_white_v2.pdf
Thought Questions about Chapter 8 due 
12 Mar 29 Chapter 9 continued, projects, & average reward RL:  Download Projects-AverageReward.pdf
 
Mar 31

Policy gradient methods:  Download AverageReward-PG.pdf

Thought Questions about Chapter 9 due
13 Apr 5

Advanced topics (Off-policy gradient TD):  Download ofPolicy-GTD.pdf

Apr 7

 

Advanced topics (Off-policy gradient TD cont...):  Download ofPolicy-GTD2.pdf

 

Assignment #4 due

Assignment #5 released (RL and function approximation)

 

14 Apr 12 Advanced topics (Least squares TD):  Download LSTD.pdf

 

Apr 14 Research project discussion

 

15 Apr 19 RL and Psychology:  Download animalLearning.pdf
Apr 21 Chapter 11: Case studies & Review:  Download summary.pdf
16 Apr 26 No class

 

Assignment #5 due

Apr 28

No class

Office hours still on

 

17 May 3

Office hours still on

May 5

No exam

Research projects due