Welcome to the TicTacToe RL AI project!
This project uses Q-learning with a lookup table to create an AI agent that can play TicTacToe at an expert level. The AI learned the optimal strategy in a matter of seconds, making it a powerful tool for anyone interested in studying reinforcement learning.
The codebase for this project is written in Python and uses the popular reinforcement learning library, TensorFlow, along with NumPy and Pandas for data manipulation. The code is optimized for performance, ensuring that the AI can play TicTacToe efficiently.
The AI agent is trained using Q-learning, which is a widely used reinforcement learning algorithm. The lookup table is used to store the Q-values for each state-action pair, allowing the AI agent to learn the optimal strategy for playing TicTacToe. The AI agent is trained by playing against itself, allowing it to learn from its own mistakes and improve its strategy over time.
The optimal strategy learned by the AI agent is based on the traditional TicTacToe game rules. The AI agent can recognize winning, losing, and draw situations and take appropriate actions to maximize its chances of winning the game.