An evaluation of fast multi-layer perceptron training techniques for games

Despite the rise of Artificial Intelligence (AI) in games leading to the adoption of many academic techniques, multi-layer perceptron (MLP) neural networks have bucked this trend and have rarely been used in a game scenario. This is normally due to long training and development times using the standard error back propagation (EBP) training technique. The purpose of this investigation was to compare alternative training techniques to EBP in order to see if they can be used to promote the use of MLP in games.
The application created to serve this purpose was a 2D top down racing game with three different training techniques to control the AI, including EBP, resilient propagation (RPROP) and Random-Minimum Bit Distance Gram Schmidt (RMGS), in which, each training technique was put through three tests.
Through these tests, it was shown that alternative training techniques, although not as accurate as EBP, reduce the training time drastically. The tests also concluded that in a racing game scenario the alternative techniques could also compete with EBP, with the RMGS training technique being the best in every test except accuracy.
This project has shown that MLP could easily be utilised in game scenarios using these alternative methods and would not require the lengthy training times of EBP.