https://github.com/comtalyst/minesweepers
Using Neural Networks and Genetic Algorithms to control movements of the bots in 2D plane
https://github.com/comtalyst/minesweepers
Last synced: 3 months ago
JSON representation
Using Neural Networks and Genetic Algorithms to control movements of the bots in 2D plane
- Host: GitHub
- URL: https://github.com/comtalyst/minesweepers
- Owner: comtalyst
- Created: 2019-10-30T21:18:31.000Z (over 6 years ago)
- Default Branch: master
- Last Pushed: 2019-10-30T22:01:34.000Z (over 6 years ago)
- Last Synced: 2025-02-24T16:45:52.777Z (over 1 year ago)
- Language: Java
- Homepage:
- Size: 15.6 KB
- Stars: 0
- Watchers: 2
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Minesweepers
The technical objective of the project is to train the "Minesweeper" bots to correctly walk through the "good mines" and evade the "bad mines" presented on a 2D plane using Genetic Algorithms and Neural Networks.
Each agent (bot) will have a unique model of Neural Network which will decide its next movement based on nearest mines.
These agents will also simulate genetic algorithms in order to select only high-fitness models to be passed to the next generation. (That's why we have multiple bots).
This project is part of the Independent Study in Advanced Programming course at George School.
Some parts of the codes were provided by the instructor, Travis.
Note that the codes have been kept in the original form at the time of submission.