Overview

This lecture note volume is mainly about the recent development that connected neural network modeling to the theoretical physics of disordered systems. It gives a detailed account of the (Little-) Hopfield model and its ramifications concerning non-orthogonal and hierarchical patterns, short-term memory, time sequences, and dynamical learning algorithms. It also offers a brief introduction to computation in layered feed-forward networks, trained by back-propagation and other methods. Kohonen's self-organizing feature map algorithm is discussed in detail as a physical ordering process. The book offers a minimum complexity guide through the often cumbersome theories developed around the Hopfield model. The physical model for the Kohonen self-organizing feature map algorithm is new, enabling the reader to better understand how and why this fascinating and somewhat mysterious tool works.Contents: Associative Memory and AttractorsThe Neuron and Its ModelHebbian LearningThe Hopfield ModelDescendants of the Hopfield ModelPatient LearningDynamics of RetrievalFeed-Forward NetworksThe Boltzmann MachineSelf-Organized Feature Maps: The Kohonen ModelAppendices: The Replica Trick Way to Mean FieldDynamics of a Two-Pattern NetworkReadership: Neuroscientists, physicists, biologists, mathematicians and computer scientists.

Full Product Details

9781299462601

Table of Contents

Reviews

Author Information

Tab Content 6

Countries Available