In the Image of the Brain : Breaking the Barrier Between the Human Mind and Intelligent Machines

Shows how a colorful cast of scientists is using their expanding knowledge of the brain and its evolution to make machines sense and think like human beings

Neural networks are computing devices capable of learning and evolving. Modeled after the neurobiology of the human brain, with immense numbers of processors, they are either simulated in computer programs or actually built out of silicon. These "brain-like" gizmos can identify a person's gender from a facial image, assess mortgage risks or locate an animal from the sound it makes. In a solid, valuable report, former Venture magazine editor Jubak takes readers to the cutting edge of this field by interviewing neural network researchers at their drawing-boards. Among them are University of Colorado's Micahael Mozer, inventor of a "Bach machine" that composes music; Stanford neuroscientist Eric Knudsen, who studies barn owls to learn how brains map spatial patterns; and Charles Gray, a researcher in Frankfurt, Germany, who believes he may have found consciousness itself in the oscillations of brain neurons firing in synchronization. In the end, though, the mind seems as elusive as ever.
Copyright 1992 Reed Business Information, Inc.

All of a sudden comes a spate of books glorifying neural networks. Do we sense a paradigm shift here? Down with the old reductionist approach of artificial intelligence? Up with the biologically more relevant parallel processing network models? So it would seem, but don't expect consensus. Unlike other recent writers in the field who argue for a specific theory (e.g., Gerald M. Edelman in Bright Air, Brilliant Fire, p. 297), Jubak (former editor of Venture magazine) provides a broad survey of current work in academia and industry, leaving it up to the reader to judge. Jubak is an enthusiast who's on top of developments in speech and pattern perception, robotics, organizing principles in brain development, and so on--but it's just not easy to convey the structure and behavior of computer networks in which the builders themselves are uncertain about what happens in the ``hidden'' layers connected to an input layer (responding to light signals, for example) and to the output layer (identifying a letter or other pattern). Overall, the models try to emulate features of the human brain in its connectivity and its ability to learn. ``Learning'' is often defined in terms of the Hebb synapse--a strengthening of the connection between neurons that fire together. Some basis for the Hebb synapse is revealed late in the book in the discovery of the NMDA receptor--one of two kinds of receptors at neuronal synapses that may be responsible for long-term potentiation. At this microcosmic level of brain science, we learn, the nerve cell itself may be a master microprocessor computing its behavior from multiple inputs summed over time and space and subject to its own feedback circuits. Jubak's useful if demanding survey reveals that the state of the science is such that the more we know the less we know; but that what the brain does is absolutely thrilling--and beautiful. (Line drawings throughout.) -- Copyright ©1992, Kirkus Associates, LP. All rights reserved.

Although the barrier between the human mind and any intelligent machine isn't close to being broken, this book reviews some fascinating projects doing cutting-edge development in neural networks. Neural networks are attempts to have computers operate by mimicking human evolution instead of traditional programming rules and logic. The projects using this approach are many and varied--from Cal Tech to MIT to Bellcore to Silicon Valley. There is a great deal of disagreement within the field probably due both to the variety of scientific disciplines involved (from mathematics to neurobiology to psychology) as well as the idiosyncratic personalities of these scientists. Jubak portrays all of this well--from both a technical and entertainment standpoint. This is not a light book to read, but it is a rewarding one. For larger science collections.
- Hilary D. Burton, Lawrence Livermore National Lab, Livermore, Cal.
Copyright 1992 Reed Business Information, Inc.