A simultaneous perturbation weak derivative estimator for stochastic neural networks

Thomas Flynn () and Felisa Vázquez-Abad ()
Additional contact information
Thomas Flynn: Brookhaven National Laboratory
Felisa Vázquez-Abad: Hunter College

Computational Management Science, 2019, vol. 16, issue 4, No 9, 715-738

Abstract: Abstract In this paper we study gradient estimation for a network of nonlinear stochastic units known as the Little model. Many machine learning systems can be described as networks of homogeneous units, and the Little model is of a particularly general form, which includes as special cases several popular machine learning architectures. However, since a closed form solution for the stationary distribution is not known, gradient methods which work for similar models such as the Boltzmann machine or sigmoid belief network cannot be used. To address this we introduce a method to calculate derivatives for this system based on measure-valued differentiation and simultaneous perturbation. This extends previous works in which gradient estimation algorithm’s were presented for networks with restrictive features like symmetry or acyclic connectivity.

Date: 2019
References: View references in EconPapers View complete reference list from CitEc
Citations:

Downloads: (external link)
http://link.springer.com/10.1007/s10287-019-00357-1 Abstract (text/html)
Access to the full text of the articles in this series is restricted.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:spr:comgts:v:16:y:2019:i:4:d:10.1007_s10287-019-00357-1

Ordering information: This journal article can be ordered from
http://www.springer. ... ch/journal/10287/PS2

DOI: 10.1007/s10287-019-00357-1

Access Statistics for this article

Computational Management Science is currently edited by Ruediger Schultz

More articles in Computational Management Science from Springer
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().