A fork of the BJNE Java NES Emulator that includes a Neural Network learning algorithm based on SethBling's MarI/O Lua script to enable the emulator to learn how to play itself. (Eclipse .project file already included, but you can use any IDE)
At this time, the only supported ROMs are 'Super Mario Bros' and 'Galaga' for the NES. Support for different ROMs can be added by altering the GameDataManager class and telling it which memory addresses to suck the 'tile map' inputs from.
This is still a WIP! It does work -to an extent- but the emulator saving/loading code (added during this project) still causes corruption at times. Feel free to submit pull requests.
####Instructions Run the main function in the 'SuperBrain' class using JRE 7 32 bit. Once the emulator opens, use the 'open' button in the GUI menu to run the NES ROM of your choice. Use the keyboard to press 'Enter' and start a level. Then, use the 'Start AI' button in the GUI menu and your keyboard controls will be relinquished to the evolving neural network.
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent