Evolution, neural networks, games, and intelligence

Intelligence pertains to the ability to make appropriate decisions in light of specific goals and to adapt behavior to meet those goals in a range of environments. Mathematical games provide a framework for, studying intelligent behavior in models of real-world settings or restricted domains. The behavior of alternative strategies in these games is defined by each individual's stimulus-response mapping. Limiting these behaviors to linear functions of the environmental conditions renders the results to be little more than a facade: effective decision making in any complex environment almost always requires nonlinear stimulus-response mappings The obstacle then comes in choosing the appropriate representation and learning algorithm. Neural networks and evolutionary algorithms provide useful means for addressing these issues. This paper describes efforts to hybridize neural and evolutionary computation to learn appropriate strategies in zero- and nonzero-sum,games, including the iterated prisoner's dilemma, tic-tac-toe, and checkers. With respect to checkers, the evolutionary algorithm was able to discover a neural network that can be used to play at a near-expert level without injecting expert knowledge about how To play the game. The implications of evolutionary learning with respect to machine intelligence ale also discussed. It is argued that evolution provides the framework for explaining naturally occuring intelligent entities and can be used to design machines that are also capable of intelligent behavior.