Learning to play Pac-Man: An evolutionary, rule-based approach

Pac-Man is a well-known, real-time computer game that provides an interesting platform for research. We describe an initial approach to developing an artificial agent that replaces the human to play a simplified version of Pac-Man. The agent is specified as a simple finite state machine and ruleset. with parameters that control the probability of movement by the agent given the constraints of the maze at some instant of time. In contrast to previous approaches, the agent represents a dynamic strategy for playing Pac-Man, rather than a pre-programmed maze-solving method. The agent adaptively "learns" through the application of population-based incremental learning (PBIL) to adjust the agents' parameters. Experimental results are presented that give insight into some of the complexities of the game, as well as highlighting the limitations and difficulties of the representation of the agent.

Alternating offer games and private rulings of the Australian Tax Office

The concept of submitting oneself to a voluntary negotiation is by no means new to big business. Formal bargaining has been quite successful over the years in providing the venue for agents to explore a more logical and mathematical approach to bargaining. However in more recent times external influences have been applied to agents who provide better deals for favored executives. This external influence has displayed itself in taxtion negotiations to the extent that tax office agents have been dismissed for irresponsible conduct. We explore this specific type of negotiation using an alternating offer bargaining game to model the particular influences, which create unfair rulings in negotiations. By the constraints of this systematic mathematical approach to negotiation, we will explore the advantages of a more formal game theoretic approach. In this presentation we will also elaborate on finding Nash Equilibrium in alternating offer games.