An optimization-based algorithm for model selection using an approximation of Akaike's Information Criterion

Rodrigo Carvajal, Gabriel Urrutia, Juan C. Agüero

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

5 Scopus citations

Abstract

In this paper, we consider an optimization approach for model selection using Akaike's Information Criterion (AIC) by incorporating the ℓ0-(pseudo)norm as a penalty function to the log-likelihood function. In order to reduce the numerical complexity of the optimization problem, we propose to approximate the ℓ0-(pseudo)norm by an exponential term. We focus on problems with hidden variables - i.e. where there are random variables that we cannot measure, and the Expectation-Maximization (EM) algorithm. We illustrate the benefits of our proposal via numerical simulations.

Original languageEnglish
Title of host publication2016 Australian Control Conference, AuCC 2016
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages217-220
Number of pages4
ISBN (Electronic)9781922107909
DOIs
StatePublished - 1 Mar 2017
Externally publishedYes
Event2016 Australian Control Conference, AuCC 2016 - Newcastle, Australia
Duration: 3 Nov 20164 Nov 2016

Publication series

Name2016 Australian Control Conference, AuCC 2016

Conference

Conference2016 Australian Control Conference, AuCC 2016
Country/TerritoryAustralia
CityNewcastle
Period3/11/164/11/16

Fingerprint

Dive into the research topics of 'An optimization-based algorithm for model selection using an approximation of Akaike's Information Criterion'. Together they form a unique fingerprint.

Cite this