Section 2: Reinforcement Learning (Stochastic Games)

Key Learning Points

• Explain the multiagent extensions of learning in MDPs.
• Learn about reinforcement learning in zero-sum stochastic games.

Activities

1. Read section 7.4 of the textbook;
2. Watch the following video(s):
   1. Stochastic Games
   2. Stochastic Games: Georgia Tech—Machine Learning
   3. Zero-Sum Stochastic Games: Georgia Tech—Machine Learning
   4. Zero-Sum Stochastic Games Two: Georgia Tech—Machine Learning
   5. Stochastic Games and Multiagent RL: Georgia Tech—Machine Learning
3. Implement Q-learning algorithm for a simple path-finding domain with NetLogo (referring to José M. Vidal’s site.)