Online learning in commercial computer games allows computer-controlled opponents to adapt to human player tactics. For online learning to work in practice, it must be fast, effective, robust, and efficient. This paper proposes a technique called “dynamic scripting” that meets these requirements. In dynamic scripting an adaptive rule-base is used for the generation of intelligent opponents on t...

We propose a novel strategy to enable autonomous agents to negotiate concurrently with multiple, unknown opponents in realtime, over complex multi-issue domains. We formalise our strategy as an optimisation problem, in which decisions are based on probabilistic information about the opponents’ strategies acquired during negotiation. In doing so, we develop the first principled approach that ena...

Dynamic behaviour learning in the face of adversarial opponents involves a) learning a basic set of strategies, and b) tuning these strategies for the specific opponents involved. Iterative approaches to dynamic learning are often slow for large state spaces, especially since in many dynamic situations, the reward is not obvious immediately, but may need to be temporally apportioned over multip...

Poker is an interesting test-bed for artificial intelligence research. It is a game of imperfect knowledge, where multiple competing agents must deal with risk management, opponent modeling, unreliable information, and deception, much like decision-making applications in the real world. Opponent modeling is one of the most difficult problems in decision-making applications and in poker it is es...

Competitive fitness is the assessment of an individual’s fitness in the context of competition with other individuals in the evolutionary system. This commonly takes one of two forms: one-population competitive fitness, where competition is solely between individuals in the same population; and N-population competitive fitness, often termed competitive coevolution. In this paper we discuss comm...

Consider a finite stage game G that is repeated infinitely often. At each time, the players have hypotheses about their opponents' repeated game strategies. They frequently test their hypotheses against the opponents' recent actions. When a hypothesis fails a test, a new one is adopted. Play is almost rational in the sense that, at each point in time, the players' strategies are ε-best replies ...

In this paper we examine the application of temporal difference methods in learning a linear state value function approximation in a game of give-away checkers. Empirical results show that the TD(λ) algorithm can be successfully used to improve playing policy quality in this domain. Training games with strong and random opponents were considered. Results show that learning only on negative game...