Simultaneous Perturbation Stochastic Approximation for Clustering of a Gaussian Mixture Model under Unknown but Bounded Disturbances

Andrei Boiarov, Oleg Granichin, Hou Wenguang

Research output: Chapter in Book/Report/Conference proceedingConference contributionResearchpeer-review

1 Scopus citations

Abstract

Multidimensional optimization holds a central role in many machine learning problems. When a model quality functional is measured with an almost arbitrary external noise, it makes sense to use randomized optimization techniques. This paper deals with the problem of clustering of a Gaussian mixture model under unknown but bounded disturbances. We introduce a stochastic approximation algorithm with randomly perturbed input (like SPSA) to solve this problem. The proposed method is appropriate for the online learning with streaming data, and it has a high speed of convergence. We study the conditions of the SPSA clustering algorithm applicability and show illustrative examples.

Original languageEnglish
Title of host publication2017 IEEE CONFERENCE ON CONTROL TECHNOLOGY AND APPLICATIONS (CCTA 2017)
PublisherIEEE Canada
Pages1740-1745
Number of pages6
StatePublished - 2017
Event1st Annual IEEE Conference on Control Technology and Applications - Hawaii, United States
Duration: 27 Aug 201730 Aug 2017

Conference

Conference1st Annual IEEE Conference on Control Technology and Applications
CountryUnited States
Period27/08/1730/08/17

Keywords

  • Clustering
  • Gaussian mixture model
  • randomized algorithm
  • SPSA
  • unknown but bounded disturbances
  • ALGORITHM
  • INPUT

Fingerprint

Dive into the research topics of 'Simultaneous Perturbation Stochastic Approximation for Clustering of a Gaussian Mixture Model under Unknown but Bounded Disturbances'. Together they form a unique fingerprint.

Cite this