Tabular methods for reinforcement learning
-
Updated
Jul 3, 2020 - Python
Tabular methods for reinforcement learning
This repo implements Deep Q-Network (DQN) for solving the Cliff Walking v0 environment of the Gymnasium library using Python 3.8 and PyTorch 2.0.1 with the finest tuning.
This project utilizes Markov Decision Process (MDP) principles to implement a custom "CliffWalking" environment in Gym, employing policy iteration to find an optimal policy for agent navigation.
Use cliff walking to compare the difference between Q-learning and SARSA algorithms in Reinforcement Learning
Add a description, image, and links to the cliffwalking topic page so that developers can more easily learn about it.
To associate your repository with the cliffwalking topic, visit your repo's landing page and select "manage topics."