Library for creating multi-layered perceptrons in Java. This project was started to create a Generative Adversial Network using multi-layer perceptrons as the classifier and generator.
This GAN is being built to trick the cheat detection system in a game called Old School Runescape by generating human-like mouse data.
The process will be:
- Building a classifier that can classify game data to be human or inhuman
- Building a generator that will attempt to generate game data and feed it to the classifier
- Improving both the classifier and the generator iteratively such that the generator can generate data to which the classifier does not know what answer to give.