https://github.com/omaraflak/frozenlake-qlearning

Resolving the FrozenLake problem from OpenAI Gym.
https://github.com/omaraflak/frozenlake-qlearning

frozenlake machine-learning openai-gym qlearning reinforcement-learning

Last synced: 6 months ago
JSON representation

Resolving the FrozenLake problem from OpenAI Gym.

• Host: GitHub
• URL: https://github.com/omaraflak/frozenlake-qlearning
• Owner: omaraflak
• Created: 2018-07-20T08:14:42.000Z (about 7 years ago)
• Default Branch: master
• Last Pushed: 2018-07-21T10:20:57.000Z (about 7 years ago)
• Last Synced: 2025-02-12T03:22:42.617Z (8 months ago)
• Topics: frozenlake, machine-learning, openai-gym, qlearning, reinforcement-learning
• Language: Python
• Size: 1.95 KB
• Stars: 1
• Watchers: 3
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

          
# FrozenLake QLearning

Resolving the FrozenLake problem from OpenAI Gym with QLearning.

# Help

Two tricks that made the program work :

* decrease **epsilon** only when the agent wins the game

* decrease the **learning rate** overtime