ECE 517: Reinforcement Learning in Artificial Intelligence Fall 2015

* Schedules are subject to changes, please check back frequently
 

Date

Topic/s Covered

Lecture Notes & Handouts

Reading Assignments (before class)

Homework & Projects

Thr - Aug 20

Class overview, Introduction

Lecture #1

 

Word Clouds on Interests and Expectations

* A good reference book on Probability theory)

* Matlab references [1][2]

 

 

Tue - Aug 25

Evaluative Feedback

 

Lecture #2

 

 

 

 

Thr Aug 27

Review of discrete-time probability fundamentals

 

Lecture #3

 

 

 

Homework #1

Due: September 8

Tue Sept 1

Discrete Time Markov Chains

 

Lecture #4

 

 

 

Thr Sep 3

The Reinforcement Learning Problem

 

Lecture #5

 

 

Lecture #6

 

 

 

 

Tue Sep 8

The Reinforcement Learning Problem (cont)

 

 

 

Thr Sep 10

Markov Decision Processes (MDPs) and Optimality Criterion in MDPs, SW/HW Considerations

 

Lecture #7

 

Homework #2

Due: September 24

 

Tue Sep 15

 

Finite Horizon MDP

 

Lecture #8

 

 

Thr Sep 17


Dynamic Programming, Policy Iteration, Value Iteration

 

 

 

 

 

Tue Sep 22

 

Class Cancelled

 

 

 

Thr Sep 24

Monte Carlo Methods

 

Lecture #9

 

 

Project #1

Due: Oct 8

Tue Sept 29

Monte Carlo Methods (cont)

Temporal Difference Learning

 

Lecture #10

 

 

 

Thr Oct 1

Temporal Difference Learning (cont)

 

 

 

 

Tue Oct 6

 

In-Class Midterm MK 405

 

 

 


Thr Oct 8

Temporal Difference Learning (cont)

Lecture #11

 

 

 

 

 

Tue Oct 13

Actor-Critic Model,

Eligibility Traces

 

 

 

Lecture #12

 

 

Homework #3

Due: Oct 22

Thr Oct 15

Fall break No Class

 

 

 

 

Tue Oct 20

Eligibility traces (cont)

 

 

Project #2

Due: Nov 3

Thr Oct 22

Intro to Neural Networks

Lecture #13

 

 

 

Tue Oct 27

Planning and Learning

Lecture #14

 

 

Homework #4

Due: Nov 5

Thr Oct 29

Case studies in RL

Lecture #15

 

 

List of suggested topics for final project

Tue Nov 3

Recurrent Neural Networks,

LSTMs

Lecture #16

 

 

Thr Nov 5

Partially Observable MDPs (POMDPs)

Lecture #17

 

 

Homework #5

Due: Nov 19

Tue Nov 10

Open topics in RL and its Applications

Lecture #18

 

 

Final project presentations schedule

Thr Nov 12

Deep reinforcement learning

 

 

 

Tue Nov 17

Project presentations

 

 

 

Wed Nov 18

3:35pm 4:25pm

Project presentations

Location: MK 525

 

 

 

Thr Nov 19

Project presentations

 

 

 

Tue Nov 24

Project presentations

 

 

 

Wed Nov 25

1:15pm 3:30pm

Project Presentations -

Tickle Engineering Bldg

Room 402

 

 

 

Thr Nov 26

Thanksgiving No Class

 

 

 

Tue Dec 11:10pm 1:45pm

Project presentations
MK 405 (meet there)
then in FH 502

 

 

 

Monday December 7

Final project reports are due

 

 

 

Matlab codes that cover most topics addressed in the course textbook can be found at: http://www.waxworksmath.com/Authors/N_Z/Sutton/sutton.html

Last Update: November 20, 2015