Showing papers by "Shaul Markovitch published in 1996"

Learning models of intelligent agents

Abstract: Agents that operate in a multi-agent system need an efficient strategy to handle their encounters with other agents involved. Searching for an optimal interactive strategy is a hard problem because it depends mostly on the behavior of the others. In this work, interaction among agents is represented as a repeated two-player game, where the agents' objective is to look for a strategy that maximizes their expected sum of rewards in the game. We assume that agents' strategies can be modeled as finite automata. A model-based approach is presented as a possible method for learning an effective interactive strategy. First, we describe how an agent should find an optimal strategy against a given model. Second, we present an unsupervised algorithm that infers a model of the opponent's automaton from its input/output behavior. A set of experiments that show the potential merit of the algorithm is reported as well.

Incorporating opponent models into adversary search

Abstract: This work presents a generalized theoretical framework that allows incorporation of opponent models into adversary search. We present the M* algorithm, a generalization of minimax that uses an arbitrary opponent model to simulate the opponent's search. The opponent model is a recursive structure consisting of the opponent's evaluation function and its model of the player. We demonstrate experimentally the potential benefit of using an opponent model. Pruning in M* is impossible in the general case. We prove a sufficient condition for pruning and present the αβ* algorithm which returns the M* value of a tree while searching only necessary branches.

Learning of resource allocation strategies for game playing

Abstract: Human chess players exhibit a large variation in the amount of time they allocate for each move. Yet, the problem of devising resource allocation strategies for game playing has not received enough attention. In this paper we present a framework for studying resource allocation strategies. We define allocation strategy and identify three major types of strategies: static, semi-dynamic, and dynamic. We then describe a method for learning semi-dynamic strategies from self-generated examples. We present an algorithm for assigning classes to the examples based on the utility of investing extra resources. The method was implemented in the domain of checkers, and experimental results show that it is able to learn strategies that improve game-playing performance.