This repository contains implementations of various game-playing agents and their corresponding game environments. The project focuses on different AI algorithms and strategies for game playing.
- Random Agent (
agents/agent_random.py) - A simple random move generator - Minimax Agent (
agents/minimax.py) - Implementation of the minimax algorithm - MCTS Agent (
agents/mcts.py) - Monte Carlo Tree Search implementation - Counterfactual Regret Agent (
agents/counterfactualregret.py) - Counterfactual Regret Minimization - Input Agent (
agents/input_agent.py) - Human input agent for testing
- Kuhn Poker (
games/kuhn.py) - Simplified poker variant - Tic-Tac-Toe (
games/tictactoe/) - Classic 3x3 game - Nocca Nocca (
games/nocca_nocca/) - Custom game implementation
- KuhnPoker.ipynb - Interactive notebook for Kuhn Poker experiments
- TicTacToe.ipynb - Interactive notebook for Tic-Tac-Toe experiments
- Nocca_Nocca.ipynb - Interactive notebook for Nocca Nocca experiments
Detailed analysis and reports are available in the following Jupyter notebooks:
- ReporteKhunPoker.ipynb - Comprehensive report on Kuhn Poker experiments and agent performance analysis
- ReporteMCTS.ipynb - Detailed report on Monte Carlo Tree Search experiments
These reports contain:
- Performance comparisons between different agents
- Algorithm analysis and optimization results
- Experimental data and visualizations
- Conclusions and insights from the experiments
This project is licensed under the terms specified in the LICENSE file.