VKohli17 / TRL_A1

Implementing multiple Reinforcement Learning algorithms from scratch to find optimal policies for OpenAI Gym Environments

RL Algorithms for OpenAI Gym Environments

* Created for an assignment for Topics in Reinforcement Learning, IIITH, Spring 2023 *

• q1.ipynb contains Value Iteration and Policy Iteration based solutions to OpenAI CliffWalking-v0.
• q2.ipynb contains solutions to OpenAI Taxy-v3 implemented using Q-Learning, SARSA, On-policy Monte Carlo and Off-Policy Monte Carlo learning.
• Both the notebooks contain graphs showing convergence of the reward function as the model approaches the optimal policy.