Emotions as Durative Dynamic State for Action Selection

Human intelligence requires decades of full-time training before it can be reliably utilised in modern economies. In contrast, AI agents must be made reliable but interesting in relatively short order. Realistic emotion representations are one way to ensure that even relatively simple specifications of agent behaviour will be expressed with engaging variation, and that social and temporal contexts can be tracked and responded to appropriately. We describe here a representation system for maintaining an interacting set of durative states in order to replicate emotional control. Our model, the Dynamic Emotion Representation (DER) integrates emotional responses and keeps track of emotion intensities changing over time. The developer can specify an interacting network of emotional states with appropriate onsets, sustains and decays. The levels of these states can be used as input for action selection, including emotional expression. We present both a general representational framework an...

• Type: Conference
• Year: 2007
• Where: IJCAI
• Authors: Emmanuel Tanguy, Philip J. Willis, Joanna Bryson
• Keywords: Artificial Intelligence, Emotion Representation, Emotional, Emotional State, IJCAI 2007