Neurocomputing (A Bradford Book)

by James A. Anderson and Andras Pellionisz

0 ratings • 0 reviews • 0 shelved
Book cover for Neurocomputing

Bookhype may earn a small commission from qualifying purchases. Full disclosure.

In bringing together seminal articles on the foundations of research, the first volume of "Neurocomputing has become an established guide to the background of concepts employed in this burgeoning field. "Neurocomputing 2 collects thirty-nine articles covering network architecture, neurobiological computation, statistics and pattern classification, and problems and applications, that suggest important directions for the evolution of neurocomputing. Many of the most valuable practical ideas for neurocomputing in the 1990s may well come from work in neurobiology and computational neuroscience. The articles in this section discuss central issues in neuroscience as well as information-processing strategies for handling computation in the mammalian nervous system. Although neural networks are useful models to explain many aspects of the function of brain and mind, they have intrinsic difficulties. These are underscored in articles by Minsky and Papert, who conclude that much current work has not answered the twenty-year-old criticisms in "Perceptrons, and by Fodor and Pylyshyn, who argue that there are aspects of human cognition that cannot be explained by simple associations. The book-concludes with interesting, diverse applications of neural networks - in digit recognition, speech recognition, object motion, arithmetic teaming, and in formal techniques for predicting the behavior of the stock market, and for predicting secondary protein structure. James A. Anderson is Professor in the Department of Cognitive and Linguistic Sciences at Brown University. Andras Pellionicz is a Research Associate Professor in the Department of Physiology and Biophysics at New York University Medical Center.Edward Rosenfeld in editor and publisher of the newsletters "Intelligence - The Future of Computing and "Medical Intelligence.
  • ISBN10 0262011190
  • ISBN13 9780262011198
  • Publish Date 1 January 1991
  • Publish Status Out of Stock
  • Out of Print 28 February 2001
  • Publish Country US
  • Publisher MIT Press Ltd
  • Imprint MIT Press
  • Format Hardcover
  • Pages 758
  • Language English