GitHub - devadigapratham/Daisy: Trying to build AlphaZero from scratch. Daisy is a game-playing algorithm that uses artificial intelligence and machine learning techniques to learn how to play board games at a superhuman level.

Project Overview

Objective: Build AlphaZero, a game-playing algorithm with superhuman capabilities in board games.

TicTacToe Implementation: Code the TicTacToe game for AlphaZero.
MCTS Algorithm: Implement the Monte Carlo Tree Search algorithm.
Model Development: Develop the model for AlphaZero.
AlphaMCTS Integration: Integrate Alpha version of MCTS into the project.
AlphaSelfPlay Implementation: Implement the self-play mechanism for AlphaZero.
AlphaTrain Process: Train AlphaZero using reinforcement learning.
AlphaTweaks: Fine-tune AlphaZero for improved performance.
ConnectFour Integration: Extend AlphaZero to play Connect Four.
AlphaParallel Implementation: Enhance performance through parallelization.
Evaluation Process: Evaluate the performance of AlphaZero.

Contributions and feedback are encouraged. Report issues or submit pull requests to improve the project.

This project is licensed under the MIT License. Feel free to use, modify, and share the code as per the license terms.

Happy coding and exploring the world of AlphaZero! 🚀

Name		Name	Last commit message	Last commit date
Latest commit History 11 Commits
AlphaMCTS.ipynb		AlphaMCTS.ipynb
AlphaZero2_0.ipynb		AlphaZero2_0.ipynb
AlphaZeroForConnectFour.ipynb		AlphaZeroForConnectFour.ipynb
Evaluation.ipynb		Evaluation.ipynb
LICENSE		LICENSE
Model.ipynb		Model.ipynb
Parallel.ipynb		Parallel.ipynb
README.md		README.md
SelfPlayAlphaModule.ipynb		SelfPlayAlphaModule.ipynb
Train.ipynb		Train.ipynb
Tweaks.ipynb		Tweaks.ipynb
model_2.pt		model_2.pt
model_7_ConnectFour.pt		model_7_ConnectFour.pt