C13
MDPs, Dynamic Programming, and TD Learning
4 lessons3 hoursPlannedFree
Outcome
Solve small decision problems and implement temporal-difference learning.
Prerequisites
- Python basics.
Lessons
- L01MDPs and returns · coming soonExercise: Define a gridworld MDP.
- L02Bellman equations and value iteration · coming soonExercise: Solve the gridworld.
- L03Monte Carlo methods · coming soonExercise: Estimate values by simulation.
- L04Q-learning and SARSA · coming soonExercise: Train an agent on a toy environment.