A Comparison of New and Old Algorithms for a Mixture Estimation Problem

We investigate the problem of estimating the proportion vector which maximizes the likelihood of a given sample for a mixture of given densities. We adapt a framework developed for supervised learning and give simple derivations for many of the standard iterative algorithms like gradient projection...

Celý popis

Uložené v:
Podrobná bibliografia
Vydané v:Machine learning Ročník 27; číslo 1; s. 97 - 119
Hlavní autori: Helmbold, David P., Schapire, Robert E., Singer, Yoram, Warmuth, Manfred K.
Médium: Journal Article
Jazyk:English
Vydavateľské údaje: Dordrecht Springer Nature B.V 1997
Predmet:
ISSN:0885-6125, 1573-0565
On-line prístup:Získať plný text
Tagy: Pridať tag
Žiadne tagy, Buďte prvý, kto otaguje tento záznam!
Popis
Shrnutí:We investigate the problem of estimating the proportion vector which maximizes the likelihood of a given sample for a mixture of given densities. We adapt a framework developed for supervised learning and give simple derivations for many of the standard iterative algorithms like gradient projection and EM. In this framework, the distance between the new and old proportion vectors is used as a penalty term. The square distance leads to the gradient projection update, and the relative entropy to a new update which we call the exponentiated gradient update (EG^sub ^). Curiously, when a second order Taylor expansion of the relative entropy is used, we arrive at an update EM^sub ^ which, for =1, gives the usual EM update. Experimentally, both the EM^sub ^-update and the EG^sub ^-update for > 1 outperform the EM algorithm and its variants. We also prove a polynomial bound on the rate of convergence of the EG^sub ^ algorithm.[PUBLICATION ABSTRACT]
Bibliografia:SourceType-Scholarly Journals-1
ObjectType-Feature-1
content type line 14
ObjectType-Article-2
content type line 23
ISSN:0885-6125
1573-0565
DOI:10.1023/A:1007301011561