RL

Tic Tac Toe with Reinforcement Learning (Value Iteration)

This is a simple Tic Tac Toe game built using C++ and SFML for the GUI.
It includes a basic AI agent trained using Reinforcement Learning (Value Iteration).

• The AI learns optimal moves by simulating all possible game states.
• It assumes the opponent plays optimally (minimax-style reasoning).
• For its first move, the AI follows an epsilon-greedy strategy to add some randomness.

How to Play

• Run the executable
• Click to place your move (you play as 'O', AI plays as 'X')
• Press R to restart the game after it's over

Requirements

• SFML library (used for graphics and input)
• C++ compiler (e.g., g++, Visual Studio)