An integrated framework for adaptive reasoning about conversation patterns

We present an integrated approach for reasoning about and learning conversation patterns in multiagent communication. The approach is based on the assumption that information about the communication language and protocols available in a multiagent system is provided in the form of dialogue sequence patterns, possibly tagged with logical conditions and instance information. We describe an integrated social reasoning architecture that is capable of (i) processing such patterns, (ii) making communication decisions in a boundedly rational way, and (iii) learning patterns and their strategic application from observation. Our method combines decision-theoretic utility maximisation, case-based reasoning methods, hierarchical reinforcement learning and cluster validation techniques to yield a comprehensive model of communicative decisionmaking and learning that can be practically implemented. The adequacy of the approach is illustrated through experimental results in complex negotiation scena...

• Type: Conference
• Year: 2005
• Where: ATAL
• Authors: Michael Rovatsos, Felix A. Fischer, Gerhard Weiß
• Keywords: ATAL 2005, Dialogue Sequence Patterns, Multiagent Communication, Reasoning