Sunday, January 08, 2006
Novamente uses BOA for AI machine
According to Notes on Minds, a company called Novamente uses a Bayesian optimization algorithm (see related post here) as part of its larger artificial intelligence architecture:
The basic idea is to dig for patterns using a bayesian optimization algorithm (a heuristically improved genetic algorithm). Patterns are cloned, mutated, and tested. Good patterns are kept and associated with weights. These weights control how much "attention" each pattern later receives.Novamente was founded by an interesting fellow I might a couple years ago named Ben Goertzel. Ben has a roving and interesting mind, and although his ambitions are large, I wouldn't bet against his efforts.
Later, patterns are connected into a hypergraph, which is a graph, but with special edges that may point to more than two vertices, and be of various types. One edge type indicates that a vertex A is a special case of vertex B; another says that several vertices are similar.