Cargando…

Pullback Bundles and the Geometry of Learning

Explainable Artificial Intelligence (XAI) and acceptable artificial intelligence are active topics of research in machine learning. For critical applications, being able to prove or at least to ensure with a high probability the correctness of algorithms is of utmost importance. In practice, however...

Descripción completa

Detalles Bibliográficos
Autor principal: Puechmorel, Stéphane
Formato: Online Artículo Texto
Lenguaje:English
Publicado: MDPI 2023
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10606266/
https://www.ncbi.nlm.nih.gov/pubmed/37895571
http://dx.doi.org/10.3390/e25101450
_version_ 1785127275409178624
author Puechmorel, Stéphane
author_facet Puechmorel, Stéphane
author_sort Puechmorel, Stéphane
collection PubMed
description Explainable Artificial Intelligence (XAI) and acceptable artificial intelligence are active topics of research in machine learning. For critical applications, being able to prove or at least to ensure with a high probability the correctness of algorithms is of utmost importance. In practice, however, few theoretical tools are known that can be used for this purpose. Using the Fisher Information Metric (FIM) on the output space yields interesting indicators in both the input and parameter spaces, but the underlying geometry is not yet fully understood. In this work, an approach based on the pullback bundle, a well-known trick for describing bundle morphisms, is introduced and applied to the encoder–decoder block. With constant rank hypothesis on the derivative of the network with respect to its inputs, a description of its behavior is obtained. Further generalization is gained through the introduction of the pullback generalized bundle that takes into account the sensitivity with respect to weights.
format Online
Article
Text
id pubmed-10606266
institution National Center for Biotechnology Information
language English
publishDate 2023
publisher MDPI
record_format MEDLINE/PubMed
spelling pubmed-106062662023-10-28 Pullback Bundles and the Geometry of Learning Puechmorel, Stéphane Entropy (Basel) Article Explainable Artificial Intelligence (XAI) and acceptable artificial intelligence are active topics of research in machine learning. For critical applications, being able to prove or at least to ensure with a high probability the correctness of algorithms is of utmost importance. In practice, however, few theoretical tools are known that can be used for this purpose. Using the Fisher Information Metric (FIM) on the output space yields interesting indicators in both the input and parameter spaces, but the underlying geometry is not yet fully understood. In this work, an approach based on the pullback bundle, a well-known trick for describing bundle morphisms, is introduced and applied to the encoder–decoder block. With constant rank hypothesis on the derivative of the network with respect to its inputs, a description of its behavior is obtained. Further generalization is gained through the introduction of the pullback generalized bundle that takes into account the sensitivity with respect to weights. MDPI 2023-10-15 /pmc/articles/PMC10606266/ /pubmed/37895571 http://dx.doi.org/10.3390/e25101450 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
Puechmorel, Stéphane
Pullback Bundles and the Geometry of Learning
title Pullback Bundles and the Geometry of Learning
title_full Pullback Bundles and the Geometry of Learning
title_fullStr Pullback Bundles and the Geometry of Learning
title_full_unstemmed Pullback Bundles and the Geometry of Learning
title_short Pullback Bundles and the Geometry of Learning
title_sort pullback bundles and the geometry of learning
topic Article
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10606266/
https://www.ncbi.nlm.nih.gov/pubmed/37895571
http://dx.doi.org/10.3390/e25101450
work_keys_str_mv AT puechmorelstephane pullbackbundlesandthegeometryoflearning