Skip to content
Ahmed Haroon
← Books

Solved Exercises for Reinforcement Learning

Chapter-by-chapter exercises

  1. 1IntroductionTic-tac-toe, self-play, symmetries and the cost of pure greed.
  2. 2Multi-armed BanditsAction-value methods, non-stationary bandits, UCB, and gradient bandits.
  3. 3Finite Markov Decision ProcessesReturns, value functions, Bellman equations, and optimal policies.
  4. 4Dynamic ProgrammingPolicy evaluation, policy iteration, value iteration, and the gambler's problem.
  5. 5Monte Carlo MethodsPrediction, control, importance sampling, blackjack, and racetrack experiments.
  6. 6Temporal-Difference LearningTD prediction, SARSA, Q-learning, Expected SARSA, and windy gridworld.