
As an Amazon Associate and affiliate partner, Menrva Books earns from qualifying purchases. Learn more
Feed-Forward Neural Networks: Vector Decomposition Analysis, Modelling and Analog Implementation presents a novel method for the mathematical analysis of neural networks that learn according to the back-propagation algorithm. The book also discusses some other recent alternative algorithms for hardware implemented perception-like neural networks. The method permits a simple analysis of the learning behaviour of neural networks, allowing specifications for their building blocks to be readily obtained. Starting with the derivation of a specification and ending with its hardware implementation, analog hard-wired, feed-forward neural networks with on-chip back-propagation learning are designed in their entirety. On-chip learning is necessary in circumstances where fixed weight configurations cannot be used. It is also useful for the elimination of most mis-matches and parameter tolerances that occur in hard-wired neural network chips. Fully analog neural networks have several advantages over other implementations: low chip area, low power consumption, and high speed operation. Feed-Forward Neural Networks is an excellent source of reference and may be used as a text for advanced courses.
This book introduces a novel mathematical method for analyzing feed-forward neural networks that utilize the back-propagation algorithm. The author, Anne-Johan Annema, presents a comprehensive approach that allows for straightforward analysis of neural network learning behavior and facilitates the direct derivation of specifications for their constituent components. The work details the entire process from specification to hardware implementation, focusing on analog hard-wired, feed-forward neural networks capable of on-chip back-propagation learning, which is crucial for scenarios where fixed weight configurations are not viable or for mitigating mismatches in hardware implementations.
The book is presented as an excellent source of reference and suitable for advanced courses, suggesting its utility for both researchers and students in the field of neural networks and hardware implementation. The detailed methodology for analyzing learning behavior and deriving specifications points to a rigorous, technical approach. The focus on analog implementations highlights practical considerations for efficiency and speed in neural network hardware.
Page Count:
256
Publication Date:
1995-01-01
ISBN-10:
1461359902
ISBN-13:
9781461359906
No comments yet. Be the first to share your thoughts!