reinforcement_learning

projets for reinforcement learning class

contains:

HW1: first homework (Reinforcement Learning Basics)

Do-it-yourself implementation of value iteration, policy iteration

HW2: second homework (Multi-Armed Bandit)

Do-it-yourself implementation of Multi-Armed Bandit algorithms : UCB (upper confidence bound), greedy policy...

HW3: second homework (Reinforcement Learning)

Do-it-yourself on :

• On-Policy Reinforcement Learning with Parametric Policy
• Off-Policy Reinforcement Learning with Value Function Approximation