Massively Parallel Artificial Intelligence: (American Association for Artificial Intelligence)

The increased sophistication and availability of massively parallel supercomputers has had two major impacts on research in artificial intelligence, both of which are addressed in this collection of new AI theories and experiments. Massively parallel computers have been used to push forward research in traditional AI topics such as vision, search, and speech. More important, these machines allow AI to expand in exciting new ways by taking advantage of research in neuroscience and developing new models and paradigms, among them associate memory, neural networks, genetic algorithms, artificial life, society-of-mind models, and subsumption architectures. A number of chapters show that massively parallel computing enables AI researchers to handle significantly larger amounts of data in real time, which changes the way that AI systems can be built, which in turn makes memory-based reasoning and neural-network-based vision systems become practical. Other chapters present the contrasting view that massively parallel computing provides a platform to model and build intelligent systems by simulating the (massively parallel) processes that occur in nature.

Table of Contents:
The challenge of massive parallelism, Hiroaki Kitano; massively parallel matching of knowledge structures, William A. Andersen et al; advanced update operations in massively parallel knowledge representation, James Geller; selecting salient features for machine learning from large candidate pools through parallel decision-tree construction, Kevin J. Cherkauer and Jude W. Shavlik; a parallel computational model for integrated speech and natural language understanding, Sang-Hwa Chung et al; example-based translation and its MIMD implementation, Satoshi Sato; language learning via perceptual/motor association - a massively parallel model, Valeriy I. Nenov and Michael G. Dyer; massively parallel search for the interpretation of aerial images, Larry Davis and P.J. Narayanan; massively parallel, adaptive, colour image processing for autonomous road following, Todd Jochem and Shumeet Baluja; BioLand - a massively parallel simulation environment for evolving distributed forms of intelligent behaviour, Gregory M. Werner and Michael G. Dyer; wafer-scale integration for massively parallel AI, Moritoshi Yasunaga and Hiroaki Kitano; evolvable hardware with genetic learning, Tetsuya Higuchi et al.

About the Author :
Hiroaki Kitano is Director of the ERATO Kitano Symbiotic Systems Project of the Japan Science and Technology Corporation and a Senior Researcher at Sony Computer Science Laboratories, Inc. James Hendler is Director of Semantic Web and Agent Technology, Maryland Information and Network Dynamics Laboratory, University of Maryland.