Examples

This page is work in progress, and contains an overview of the documentation for selected models and environments used in the course.

• The Pacman Game
• Control models
• Week 1: The Pacman game
• Week 1: The Inventory-control game
• Week 2: Optimal planning in the Inventory-environment
• Week 2: Optimal planning with Pacman
• Week 3: Frozen lake and dynamical programming
• Week 3: Harmonic Oscillator
• Week 3: Pendulum with random actions
• Week 4: PID Control
• Week 8: Simple bandit
• Week 8: UCB bandit algorithm
• Week 9: Policy evaluation
• Week 9: Policy iteration
• Week 9: Value iteration
• Week 10: MC Control
• Week 10: TD-learning
• Week 10: MC value estimation
• Week 11: Sarsa
• Week 11: Q-learning
• Week 11: N-step sarsa
• Week 11: Mountain-car with linear feature approximators
• Week 12: TD(Lambda)
• Week 12: Sarsa(Lambda)
• Week 13: DynaQ