Cargando…

Improving the Performance and Stability of TIC and ICE

Takeuchi’s Information Criterion (TIC) was introduced as a generalization of Akaike’s Information Criterion (AIC) in 1976. Though TIC avoids many of AIC’s strict requirements and assumptions, it is only rarely used. One of the reasons for this is that the trace term introduced in TIC is numerically...

Descripción completa

Detalles Bibliográficos
Autor principal: Ward, Tyler
Formato: Online Artículo Texto
Lenguaje:English
Publicado: MDPI 2023
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10048287/
https://www.ncbi.nlm.nih.gov/pubmed/36981400
http://dx.doi.org/10.3390/e25030512
_version_ 1785014145837432832
author Ward, Tyler
author_facet Ward, Tyler
author_sort Ward, Tyler
collection PubMed
description Takeuchi’s Information Criterion (TIC) was introduced as a generalization of Akaike’s Information Criterion (AIC) in 1976. Though TIC avoids many of AIC’s strict requirements and assumptions, it is only rarely used. One of the reasons for this is that the trace term introduced in TIC is numerically unstable and computationally expensive to compute. An extension of TIC called ICE was published in 2021, which allows this trace term to be used for model fitting (where it was primarily compared to L2 regularization) instead of just model selection. That paper also examined numerically stable and computationally efficient approximations that could be applied to TIC or ICE, but these approximations were only examined on small synthetic models. This paper applies and extends these approximations to larger models on real datasets for both TIC and ICE. This work shows the practical models may use TIC and ICE in a numerically stable way to achieve superior results at a reasonable computational cost.
format Online
Article
Text
id pubmed-10048287
institution National Center for Biotechnology Information
language English
publishDate 2023
publisher MDPI
record_format MEDLINE/PubMed
spelling pubmed-100482872023-03-29 Improving the Performance and Stability of TIC and ICE Ward, Tyler Entropy (Basel) Article Takeuchi’s Information Criterion (TIC) was introduced as a generalization of Akaike’s Information Criterion (AIC) in 1976. Though TIC avoids many of AIC’s strict requirements and assumptions, it is only rarely used. One of the reasons for this is that the trace term introduced in TIC is numerically unstable and computationally expensive to compute. An extension of TIC called ICE was published in 2021, which allows this trace term to be used for model fitting (where it was primarily compared to L2 regularization) instead of just model selection. That paper also examined numerically stable and computationally efficient approximations that could be applied to TIC or ICE, but these approximations were only examined on small synthetic models. This paper applies and extends these approximations to larger models on real datasets for both TIC and ICE. This work shows the practical models may use TIC and ICE in a numerically stable way to achieve superior results at a reasonable computational cost. MDPI 2023-03-16 /pmc/articles/PMC10048287/ /pubmed/36981400 http://dx.doi.org/10.3390/e25030512 Text en © 2023 by the author. https://creativecommons.org/licenses/by/4.0/Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
spellingShingle Article
Ward, Tyler
Improving the Performance and Stability of TIC and ICE
title Improving the Performance and Stability of TIC and ICE
title_full Improving the Performance and Stability of TIC and ICE
title_fullStr Improving the Performance and Stability of TIC and ICE
title_full_unstemmed Improving the Performance and Stability of TIC and ICE
title_short Improving the Performance and Stability of TIC and ICE
title_sort improving the performance and stability of tic and ice
topic Article
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10048287/
https://www.ncbi.nlm.nih.gov/pubmed/36981400
http://dx.doi.org/10.3390/e25030512
work_keys_str_mv AT wardtyler improvingtheperformanceandstabilityofticandice