TY - GEN

T1 - Simultaneous Perturbation Stochastic Approximation for Clustering of a Gaussian Mixture Model under Unknown but Bounded Disturbances

AU - Boiarov, Andrei

AU - Granichin, Oleg

AU - Hou Wenguang, null

PY - 2017

Y1 - 2017

N2 - 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.

AB - 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.

KW - Clustering

KW - Gaussian mixture model

KW - randomized algorithm

KW - SPSA

KW - unknown but bounded disturbances

KW - ALGORITHM

KW - INPUT

M3 - статья в сборнике материалов конференции

SP - 1740

EP - 1745

BT - 2017 IEEE CONFERENCE ON CONTROL TECHNOLOGY AND APPLICATIONS (CCTA 2017)

PB - IEEE Canada

Y2 - 27 August 2017 through 30 August 2017

ER -