CS 371 - Introduction to Artificial Intelligence Homework #6

Due: All parts due at the beginning of class 19.
Note: Work is to be done in pairs. 

Q-Learning

Implement and demonstrate a successful application of the Q-learning (section 6.5, of Reinforcement Learning: an introduction) problem for a simple problem from the following list:

• the Mountain Car problem (example 10.1)
• Blackjack (Example 5.1) / Approach n
• Cliff Walking (Example 6.6) / other Grid World problems, or
• a simple pre-approved problem of your choice.

In class, be prepared to informally present your work in 5-7 minutes:

• defining the problem with associated states, actions, and rewards,
• showing your use of the Q-Learning update, and
• demonstrating successful learning according to a well-defined metric.