Classic arcade games where the AI opponent learns and adapts using Q-Learning (Reinforcement Learning).
- Pong — AI paddle learns your shot patterns and adapts its strategy
- Snake — AI snake evolves across generations, getting smarter each time
The AI uses Q-Learning — a model-free reinforcement learning algorithm. It starts with zero knowledge and learns purely from experience:
- Reward for scoring / eating food
- Penalty for losing / crashing
- Q-Table maps game states → optimal actions
- Epsilon-greedy exploration decays over time
- Live Q-Value visualization — see what the AI is thinking in real time
- AI generation tracking (Snake)
- Pattern detection — AI identifies your play style
- Difficulty levels
pip install -r requirements.txt
streamlit run app.pyAjith Raj — github.com/ajithhraj
MIT