Welcome to Snake AI! This simulation will train an AI to play Snake using Q-learning. The AI will learn through trial and error, improving its strategy over time.
How it works:
- Exploration Rate starts at 100% (random moves)
- Gradually decreases to 1% (using learned strategies)
- Higher exploration = more random moves
- Lower exploration = using learned patterns
- Best score is saved across sessions