The Complexity of Learning in Neural Networks - Theory and Application

Abstract

Randornized algorithms are proposed and analyzed for learning binary weights for a neuron and links to the theory of random graphs are established. Optimal stopping phenomena in gradient descent learning algorithms are characterized and explained in terms of a time-varying effective machine complexity. The finite sample performance of the k-nearest neighbor algorithm for pattern recognition is rigorously characterized. Tradeoffs in learning from mixtures of labeled and unlabeled examples are determined.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Dec 02, 1996
Accession Number
ADA342305

Entities

People

  • Santosh S. Venkatesh

Organizations

  • Moore School of Electrical Engineering

Tags

Communities of Interest

  • Energy and Power Technologies
  • Human Systems

DTIC Thesaurus Topics

  • Algorithms
  • Computational Science
  • Computer Programming
  • Computer Simulations
  • Computers
  • Electrical Engineering
  • Information Processing
  • Information Science
  • Information Systems
  • Information Theory
  • Integer Programming
  • Neural Networks
  • Pattern Recognition
  • Probability
  • Probability Distributions
  • Random Variables
  • Recognition

Fields of Study

  • Computer science

Readers

  • Neuroscience
  • Operations Research
  • Statistical inference.

Technology Areas

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