Pattern reconstruction with restricted Boltzmann machines

  • Giuseppe Genovese

    University of Zurich, Zurich, Switzerland
Pattern reconstruction with restricted Boltzmann machines cover
Download PDF

This article is published open access under our Subscribe to Open model.

Abstract

Restricted Boltzmann machines are energy models made of a visible and a hidden layer. We identify an effective energy function describing the zero-temperature landscape on the visible units and depending only on the tail behaviour of the hidden layer prior distribution. Studying the location of the local minima of such an energy function, we show that the ability of a restricted Boltzmann machine to reconstruct a random pattern depends indeed only on the tail of the hidden prior distribution. We find that hidden priors with strictly super-Gaussian tails give only a logarithmic loss in pattern retrieval, while an efficient retrieval is much harder with hidden units with strictly sub-Gaussian tails; if the hidden prior has Gaussian tails, the retrieval capability is determined by the number of hidden units (as in the Hopfield model).

Cite this article

Giuseppe Genovese, Pattern reconstruction with restricted Boltzmann machines. Math. Stat. Learn. 7 (2024), no. 3/4, pp. 155–187

DOI 10.4171/MSL/45