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

Title: An optimization-based algorithm for model selection using an approximation of Akaike's Information Criterion
Creator: Carvajal, Rodrigo; Urrutia, Gabriel; Agüero, Juan C.
Relation: 2016 Australian Control Conference (AuCC). Proceeedings of the 2016 Australian Control Conference (AuCC) (Newcastle, N.S.W. 03-04 November, 2016) p. 217-220
Publisher Link: http://dx.doi.org/10.1109/AUCC.2016.7868191
Publisher: Institute of Electrical and Electronics Engineers (IEEE)
Resource Type: conference paper
Date: 2016
Description: 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.
Subject: signal processing algorithms; approximation algorithms; computational modeling; Australia; load modeling; cost function
Identifier: http://hdl.handle.net/1959.13/1343960
Identifier: uon:29285
Identifier: ISBN:9781922107909
Language: eng
Reviewed

		Thumbnail	File	Description	Size	Format