Emotion based adaptive reasoning for resource bounded agents

In the design of resource bounded agents, high-level cognitive activities, such as reasoning, raise important problems related both to the adaptive ability and to the computational complexity of the underlying cognitive processes. To address these problems, we adopt an agent model where emotion and cognition are conceived as two integrated aspects of intelligent behavior and we present affective-emotional mechanisms that support the adaptation to changing environments and a controlled use of resources. These mechanisms produce an attention field that constrains the input to reasoning processes and also regulate the activation period of those processes. Experimental results are presented to illustrate this approach and to evaluate it by comparison with reference results concerning intention reconsideration policies. Categories and Subject Descriptors I.2.11 [Artificial Intelligence]: Distributed Artificial Intelligence - Intelligent agents. General Terms Design, Theory, Performance. Ke...

• Type: Conference
• Year: 2005
• Where: ATAL
• Authors: Luís Morgado, Graça Gaspar
• Keywords: Artificial Intelligence, ATAL 2005, Raise Important Problems, Resource Bounded Agents