Moreover, one of the main roles of such models is to serve as design guide- lines for the creation of agents, and while there is work illu strating that role in cooperative interaction, there has been no empi rical work done to validate competitive BDI models. Multiagent research provides an extensive literature on formal Belief-Desire-Intention (BDI) based models describing the notions of teamwork and cooperation, but adversarial and competitive rela- tionships have received very little formal BDI treatment. The results show that our MP-mix strategy significantly outperforms MaxN and paranoid in various settings in all three games. In addition, we also introduce the opponent impact (OI) measure, which measures the players' ability to impede their opponents' efforts, and show its relation to the relative performance of the MP-mix strategy. To evaluate our new algorithm, we performed extensive experimental evaluation on three multiplayer domains: Hearts, Risk, and Quoridor. The MP-mix algorithm examines the current situation and decides whether the root player should follow the MaxN principle, the paranoid principle, or the newly presented directed offensive principle.
We therefore suggest the MaxN-paranoid mixture (MP-Mix) algorithm: a multiplayer adversarial search that switches search strategies according to the game situation. There is no definite answer as to which approach is better, and their main shortcoming is that their strategy is fixed. In the second approach, the paranoid algorithm, the player prepares for the worst case by assuming the opponents will select the worst move with respect to him. The first approach, which stems from the MaxN algorithm, assumes each opponent will follow his highest valued heuristic move. When constructing a search tree for multiplayer games, there are two basic approaches to propagating the opponents' moves. In addition, we explore the application of our approach by analyzing log files of completed Connect Four games, and gain additional insights on the axioms’ appropriateness. We illustrate the advantages of using the model as anĪrchitectural guideline by building agents for two adversarial environments: the Connect Four game and the Risk strategic board game. To serve as design principles for building such adversarial agents. We then present behavioral axioms that are intended We define the Adversarial Activity by describing the mental states of an agent situated in such environment. implicitly modeling the opponent as an omniscient utility maximizer, rather than leveraging a more nuanced,Įxplicit opponent model). In complex environments, attempts to use classical utility-based search methods with bounded rational agents can raise a variety This paper presents the Adversarial Activity model, a formal Beliefs-Desires-Intentions (BDI) based model for bounded rational agents operating in a zero-sum environment. This form of interaction has not yet been formallyĭefined in terms of the agents mental states, beliefs, desires and intentions. However, multiagent environments are often not cooperative nor collaborative in manyĬases, agents have conflicting interests, leading to adversarial interactions. Multiagent research provides an extensive literature on formal Beliefs-Desires-Intentions (BDI) based models describing the
0 kommentar(er)
