Talking Nets: An Oral History of Neural Networks

by James A. Anderson (Editor), Edward Rosenfeld (Editor)

Since World War II, a group of scientists has been attempting to understand the human nervous system and to build computer systems that emulate the brain's abilities. Many of the early workers in this field of neural networks came from cybernetics; others came from neuroscience, physics, electrical engineering, mathematics, psychology, even economics. In this collection of interviews, those who helped to shape the field share their childhood memories, their influences, how they became interested in neural networks, and what they see as its future.The subjects tell stories that have been told, referred to, whispered about, and imagined throughout the history of the field. Together, the interviews form a Rashomon-like web of reality. Some of the mythic people responsible for the foundations of modern brain theory and cybernetics, such as Norbert Wiener, Warren McCulloch, and Frank Rosenblatt, appear prominently in the recollections. The interviewees agree about some things and disagree about more. Together, they tell the story of how science is actually done, including the false starts, and the Darwinian struggle for jobs, resources, and reputation. Although some of the interviews contain technical material, there is no actual mathematics in the book.Contributors : James A. Anderson, Michael Arbib, Gail Carpenter, Leon Cooper, Jack Cowan, Walter Freeman, Stephen Grossberg, Robert Hecht-Neilsen, Geoffrey Hinton, Teuvo Kohonen, Bart Kosko, Jerome Lettvin, Carver Mead, David Rumelhart, Terry Sejnowski, Paul Werbos, Bernard Widrow.

Genres: Artificial Intelligence, History, Science, Technology

448 pages, Paperback

First published June 23, 1998

Reviews

[Sam Johnson · Rating 3 out of 5]

"I like to ask researchers where they get their ideas. The only answer I've heard that makes sense is, 'You vary your input if you want to vary your output.' Do lots of things. If you've gotta take drugs, take drugs. Take long walks, meditate, watch a lot of movies, learn a new language, read different books, argue the other side of the debate - anything you can to vary your stimuli.
And then you have to, as they say, 'keep the ass in the seat.' You actually have to sit down and write. Do it in a disciplined way..." -Bart Kosko, Chapter 17

[Gavin · Rating 4 out of 5]

Delightful. Old men holding court; Lettvin goes for about 18 pages of monologue following one prompt. They were poor as church mice and hanging on in physiology and psychiatry and electrical engineering. They dreamed of logic and found it wanting. They were right.

Includes rare bits on the fragile genius Walter Pitts. Offered a place at Cambridge with Russell at 12. Aged 15:

[Walter Pitts] gotten hold of Carnap's new book on logic. This was in 1938. He walks into Carnap's office with his own annotated version of the book, pointing out some flaws. And he gives it to Carnap, talks to him a while, then goes out, but doesn't introduce himself. Carnap spends the next couple of months hunting high and low for that "newsboy who knew logic." In the end, he did find Walter and persuaded the University of Chicago to give him some menial job. Walter had no funds, had separated himself from his family, so that was good.

In many respects, he was like the eccentrics you read about in England. He had exactly that quality. But he was a most winning person. I mean, almost everyone who knew him was fond of him. He was in a sense almost pure thought, thought personified, but with a delightful understanding about things generally, and was a most amiable companion.

Quotes here.

[David Olmsted · Rating 4 out of 5]

This is a wonderful collection of interviews with most of the American pioneers of adaptive artificial neural networks. The only significant ones missing ones are the inventor of the Perceptron, Frank Rosenblatt, who died in 1971, and John Hopfield, inventor of the Hopfield networks in which a whole network converges to some stable state. The Japanese researchers are also not interviewed, most notably the inventor of the Cognitron, Kunihiko Fukushima.

These networks are considered artificial because they use analog numbers to represent the average action potential frequency of real neurons. Using such numbers along with regular networks also allowed mathematics to be used for some analysis. Unfortunately not much progress resulted after this pioneering stage leaving the frontier of brain theory to the more difficult spiking neural networks.