FrozenLake-game is a Python implementation of the FrozenLake environment in OpenAI's Gym, demonstrating how to apply Reinforcement Learning (RL) using Q-Learning.
This project explores the challenges of training an agent in the FrozenLake environment, specifically addressing the reward structure that affects training efficiency.
To improve training efficiency:
- The agent will receive a small reward for each step taken while on Frozen (F).
- A discount factor will be applied to encourage the agent to reach the final goal as soon as possible.
- Implemented a Q-Learning algorithm for reinforcement learning in Python.