Library for building game trees and running the minimax algorithm for turn-based games
npm install minimaxerThis library features the negamax, minimax and maxn (experimental) algorithms for find the best moves in turn based games. It can be used to build game-playing 'A.I'.
It features a number of tuning options:
- Depth based searches | Search to a specified depth (number of moves ahead).
- Iterative Deepening searches | Search 1 depth further each time until final depth reached.
- Time based searches | Search deeper and deeper until time expires.
- Alpha-beta pruning | Greatly reduces search space.
- Presorting nodes | Combines with alpha-beta pruning for even better results.
- Optimised sort | Makes presorting even faster.
- Random selections | Select moves randomly (weighted or not) to appear less 'robotic'.
and many more!
``bash``
npm i minimaxmer
See the docs at https://domw95.github.io/minimaxer/index.html.
See the series of blog posts at https://domwil.co.uk/minimaxer/.
See the examples directory for some simple implementations.
See my Azul project for a more advanced use case.