Alternating minimization and Boltzmann machine learning

  • Authors:

  • W. Byrne

  • Affiliations:

  • Dept. of Electr. Eng., Maryland Univ., College Park, MD

  • Venue:
  • IEEE Transactions on Neural Networks
  • Year:
  • 1992

Quantified Score

Hi-index 0.00

Visualization

Abstract

Training a Boltzmann machine with hidden units is appropriately treated in information geometry using the information divergence and the technique of alternating minimization. The resulting algorithm is shown to be closely related to gradient descent Boltzmann machine learning rules, and the close relationship of both to the EM algorithm is described. An iterative proportional fitting procedure for training machines without hidden units is described and incorporated into the alternating minimization algorithm