In this paper, we develop a Mean Field Games approach to Cluster Analysis. We consider a finite mixture model, given by a convex combination of probability density functions, to describe the given data set. We interpret a data point as an agent of one of the populations represented by the components of the mixture model, and we introduce a corresponding optimal control problem. In this way, we obtain a multi-population Mean Field Games system which characterizes the parameters of the finite mixture model. Our method can be interpreted as a continuous version of the classical Expectation–Maximization algorithm.

A mean field games approach to cluster analysis / Aquilanti, L.; Cacace, S.; Camilli, F.; De Maio, R.. - In: APPLIED MATHEMATICS AND OPTIMIZATION. - ISSN 0095-4616. - 84:1(2021), pp. 299-323. [10.1007/s00245-019-09646-2]

A mean field games approach to cluster analysis

Aquilanti L.;Cacace S.;Camilli F.;De Maio R.
2021

Abstract

In this paper, we develop a Mean Field Games approach to Cluster Analysis. We consider a finite mixture model, given by a convex combination of probability density functions, to describe the given data set. We interpret a data point as an agent of one of the populations represented by the components of the mixture model, and we introduce a corresponding optimal control problem. In this way, we obtain a multi-population Mean Field Games system which characterizes the parameters of the finite mixture model. Our method can be interpreted as a continuous version of the classical Expectation–Maximization algorithm.
cluster analysis; expectation–maximization algorithm; mean field games; mixture model; multi-population model
01 Pubblicazione su rivista::01a Articolo in rivista
A mean field games approach to cluster analysis / Aquilanti, L.; Cacace, S.; Camilli, F.; De Maio, R.. - In: APPLIED MATHEMATICS AND OPTIMIZATION. - ISSN 0095-4616. - 84:1(2021), pp. 299-323. [10.1007/s00245-019-09646-2]
File allegati a questo prodotto
File Dimensione Formato  
Aquilanti_Mean_2020.pdf

accesso aperto

Tipologia: Versione editoriale (versione pubblicata con il layout dell'editore)
Licenza: Tutti i diritti riservati (All rights reserved)
Dimensione 915.78 kB
Formato Adobe PDF
915.78 kB Adobe PDF Visualizza/Apri PDF

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11573/1401952
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 0
  • ???jsp.display-item.citation.isi??? 1
social impact