The objective of this study is to focus on the automatic generation of game artificial intelligence (AI) controllers for Ms. Pac-Man\r\nagent by using artificial neural network (ANN) and multiobjective artificial evolution. The Pareto Archived Evolution Strategy\r\n(PAES) is used to generate a Pareto optimal set of ANNs that optimize the conflicting objectives of maximizing Ms. Pac-Man scores\r\n(screen-capture mode) and minimizing neural network complexity.This proposed algorithm is called Pareto Archived Evolution\r\nStrategy Neural Network or PAESNet. Three different architectures of PAESNet were investigated, namely, PAESNet with fixed\r\nnumber of hidden neurons (PAESNet F), PAESNet with varied number of hidden neurons (PAESNet V), and the PAESNet with\r\nmultiobjective techniques (PAESNet M). A comparison between the single- versus multiobjective optimization is conducted in\r\nboth training and testing processes. In general, therefore, it seems that PAESNet F yielded better results in training phase. But the\r\nPAESNet M successfully reduces the runtime operation and complexity of ANN by minimizing the number of hidden neurons\r\nneeded in hidden layer and also it provides better generalization capability for controlling the game agent in a nondeterministic\r\nand dynamic environment.
Loading....