Programming Assignment 2

Due April 20

Sutton and Barto book, exercise 6.6 (windy gridworld with king moves, and stay moves). Please implement SARSA and Q-learning for this problem. Perform 30 independent runs and provide graphs similar with the one in the book, for both algorithms.

This assignment is worth 10% of your final grade

Prof. Doina PRECUP
Last modified: Mon Mar 4 15:00:59 EST 2002