Creator
Covers: theory of Model-Free Policy Evaluation
Estimated time needed to finish: 15 minutes
Questions this item addresses:
  • What is Temporal Difference (TD) policy evaluation?
How to use this item?

Go through slides 30 to 37

Author(s) / creator(s) / reference(s)
Prof. Emma Brunskill
0 comment
Recipe
publicShare
Star(0)

Model-Free Policy Evaluation

Contributors
Total time needed: ~3 hours
Objectives
Basic understanding of model-free approaches for policy evaluation
Potential Use Cases
Strategic Games, Robotics
Who is This For ?
BEGINNERAnyone who is interested in learning the concepts and real world applications of Reinforcement Learning
Click on each of the following annotated items to see details.
Resources6/6
ARTICLE 1. What are expected values, variance, and covariance?
  • What are expected values and how do these relate to the concept of covariance?
20 minutes
OTHER 2. Monte-Carlo (MC) Policy Evaluation
  • What is Monte-Carlo policy evaluation technique?
20 minutes
OTHER 3. Temporal Difference (TD) Policy Evaluation
  • What is Temporal Difference (TD) policy evaluation?
15 minutes
BOOK_CHAPTER 4. MDP, MC and TD sections from Reinforcement Learning book
  • What is Markov Decision Process (MDP)?
  • What is Monte-Carlo (MC) Learning?
  • What is Temporal Difference (TD) Learning?
30 minutes
VIDEO 5. Markov Decision Processes - Part 1
  • Definition of Markov Decision Processes?
  • What is Markov about MDPs?
  • What is V-value and Q-value?
30 minutes
VIDEO 6. Markov Decision Processes - Part 2
  • What are Bellman equations?
  • What is Value iteration?
  • What is Policy iteration?
30 minutes

Concepts Covered

0 comment