https://github.com/mahmood-anaam/reinforcement-learning-cliff-walking

Implementation of Q-learning and SARSA algorithms in the Cliff Walking environment. Explore and compare reinforcement learning techniques.
https://github.com/mahmood-anaam/reinforcement-learning-cliff-walking

cliff-walking-problem gymnasium-environment python q-learning reinforcement-learning sarsa-learning

Last synced: about 2 months ago
JSON representation

Implementation of Q-learning and SARSA algorithms in the Cliff Walking environment. Explore and compare reinforcement learning techniques.

• Host: GitHub
• URL: https://github.com/mahmood-anaam/reinforcement-learning-cliff-walking
• Owner: Mahmood-Anaam
• Created: 2024-08-25T00:48:25.000Z (9 months ago)
• Default Branch: main
• Last Pushed: 2024-08-25T01:10:13.000Z (9 months ago)
• Last Synced: 2025-02-08T21:11:37.786Z (3 months ago)
• Topics: cliff-walking-problem, gymnasium-environment, python, q-learning, reinforcement-learning, sarsa-learning
• Language: Jupyter Notebook
• Homepage:
• Size: 227 KB
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        

# Reinforcement Learning: Cliff Walking

[![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/Mahmood-Anaam/reinforcement-learning-cliff-walking/blob/main/reinforcement-learning-cliff-walking.ipynb)

## Overview

This repository contains the implementation of two fundamental reinforcement learning algorithms, **Q-learning** and **SARSA**, applied to the **Cliff Walking** environment. The project explores how these algorithms learn to navigate the gridworld, avoid the cliff, and reach the goal while minimizing penalties.

![Cliff Walking Environment](cliff_walking.gif)

## Key Concepts

- **Parts:**

  - **Q-learning:** An off-policy algorithm that learns the optimal policy by estimating the maximum future rewards.

  - **SARSA:** An on-policy algorithm that updates its policy based on the actual actions taken, leading to potentially safer but less aggressive strategies.

  - **Comparison:** A detailed comparison of the paths chosen by each algorithm, highlighting differences in exploration and exploitation behaviors.

- **Tasks:**

  - Implement and evaluate the Q-learning algorithm.

  - Implement and evaluate the SARSA algorithm.

  - Compare and analyze the optimal policies derived from both algorithms.