Analysis and Synthesis of Feedforward Neural Networks Using Discrete Affine Wavelet Transformations

Abstract

In this paper we develop a representation of a class of feedforward neural networks in terms of discrete affine wavelet transforms. It is shown that by appropriate grouping of terms, feedforward neural networks with sigmoidal activation functions can be viewed as architectures which implement affine wavelet decompositions of mappings This result follows simply from the observation that standard feedforward network architectures possess an inherent translation-dilation structure and every node implements the same activation function. It is shown that the wavelet transform formalism provides a mathematical framework within which it is possible to perform both analysis and synthesis of feedforward networks. For the purposes of analysis, the wavelet formulation characterizes a class (L(exp 2)(IR) of mappings which can be implemented by feedforward networks as well as reveals the exact implementation of a given mapping in this class. Spatio-spectral localization properties of wavelets can be exploited in synthesizing a feedforward network to perform a given approximation task. Synthesis procedures based on spatio-spectral localization result in reducing the training problem to one of convex optimization. We outline two such synthesis schemes.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Jan 01, 1990
Accession Number
ADA444558

Entities

People

  • P.S.Krishnaprasad
  • Y. C. Pati

Organizations

  • University of Maryland

Tags

DTIC Thesaurus Topics

  • Abstracts
  • Computing System Architectures
  • Information Operations
  • Network Architecture
  • Neural Networks
  • Standards
  • Universities
  • Wavelet Transforms

Fields of Study

  • Computer science

Readers

  • Adaptive Control and Estimation with Uncertainty in Dynamic Systems.
  • Image Processing and Computer Vision.
  • Neural Network Machine Learning.

Technology Areas

  • AI & ML
  • AI & ML - Machine Learning Algorithms
  • AI & ML - Neural Networks