ThomCat: A Bayesian Blackboard Model of Hierarchical Temporal Perception

We present a Bayesian blackboard system for temporal perception, applied to a minidomain task in musical scene analysis. It is similar to the classic Copycat architecture (Hofstadter 1995) but is derived from rigourous modern Bayesian network theory (Bishop 2006), with heuristics added for speed. It borrows ideas of priming, pruning and attention and fuses them with modern inference algorithms, and provides a general theory of hierarchical constructive perception, illustrated and implemented in a minidomain.

• Type: Conference
• Year: 2008
• Where: FLAIRS
• Authors: Charles W. Fox
• Keywords: Artificial Intelligence, Classic Copycat Architecture, FLAIRS 2008, Hierarchical Constructive Perception, Modern Bayesian Network