CS234: Deep Reinforcement Learning Stanford: Programming assignments solutions

This reposity contains my solutions for the programming assignments of the CS234: Reinforcement Learning Winter 2022 course offered by Stanford.

Assignment 1: Tabular MDP planning

• Assignment 1: Tabular MDP planning

Assignment 2: Deep Q-Learning

• Q4: ϵ-Greedy Exploration Strategy
• Q5: Linear Approximation
• Q6: Implementing DeepMind’s DQN
• Q7: Training DQN on Atari

Assignment 3: Policy Gradient Methods

Implementing the REINFORCE policy gradient algorithm to train a Deep Neural Network to play OpenAI Gym environments

• Policy: Code for Categorical and Gaussian Policy
• Policy gradient: Implementing the REINFORCE policy gradient algorithm.
• Baseline Network: Implementing the baseline network to use the REINFORCE algorithm with baseline