An Argument-Based Framework to Model an Agent’s Beliefs in a Dynamic Environment

Abstract

AbstractOne of the most difficult problems in multiagent systems involves representing knowledge and beliefs of agents in dynamic environments. New perceptions modify an agent’s current knowledge about the world, and consequently its beliefs. Such revision and updating process should be performed efficiently by the agent, particularly in the context of real time constraints.This paper introduces an argument-based logic programming language called Observation-based Defeasible Logic Programming (ODeLP). An ODeLP program is used to represent an agent’s knowledge in the context of a multiagent system. The beliefs of the agent are modeled with warranted goals computed on the basis of the agent’s program. New perceptions from the environment result in changes in the agent’s knowledge handled by a simple but effective updating strategy. The process of computing beliefs in a changing environment is made computationally attractive by integrating a “dialectical database” with the agent’s program, providing precompiled information about inferences. We present algorithms for creation and use of dialectical databases.KeywordsMultiagent SystemLogic ProgrammingHigh SalaryGround InstanceDefeasible LogicThese keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.