Bregman divergences for physically informed discrepancy measures for learning and computation in thermomechanics

With view on the context of convex thermomechanics, we propose tools based on the concept of Bregman divergence, a notion introduced in the 1960s and used in learning and optimization as well. This study is motivated by the need of “discrepancy measures” between physically constrained fields that ar...

Full description

Saved in:
Bibliographic Details
Main Author: Andrieux, Stéphane
Format: Article
Language:English
Published: Académie des sciences 2023-02-01
Series:Comptes Rendus. Mécanique
Subjects:
Online Access:https://comptes-rendus.academie-sciences.fr/mecanique/articles/10.5802/crmeca.164/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1825206001708564480
author Andrieux, Stéphane
author_facet Andrieux, Stéphane
author_sort Andrieux, Stéphane
collection DOAJ
description With view on the context of convex thermomechanics, we propose tools based on the concept of Bregman divergence, a notion introduced in the 1960s and used in learning and optimization as well. This study is motivated by the need of “discrepancy measures” between physically constrained fields that are used both in traditional algorithms, analysis methods and data driven modelling or applications as well.We give also a characterization of symmetrical Bregman divergences through their generating functions which can only be quadratic forms. Some properties of the Bregman divergence and the introduced concept of Bregman Gap between couples of dual quantities are given, and some existing errors in thermomechanics are recovered. Exploiting the framework of Standard Generalized Materials, we give a set of generating functions for a large range of applications, including coupled multi-physics. Finally some results useful for computational geometry are detailed.
format Article
id doaj-art-5a18708b931947639797701ea1d32c2c
institution Kabale University
issn 1873-7234
language English
publishDate 2023-02-01
publisher Académie des sciences
record_format Article
series Comptes Rendus. Mécanique
spelling doaj-art-5a18708b931947639797701ea1d32c2c2025-02-07T13:46:51ZengAcadémie des sciencesComptes Rendus. Mécanique1873-72342023-02-01351G1598110.5802/crmeca.16410.5802/crmeca.164Bregman divergences for physically informed discrepancy measures for learning and computation in thermomechanicsAndrieux, Stéphane0ONERA, Université Paris-Saclay, Chemin de la Hunière, 91123 Palaiseau, FranceWith view on the context of convex thermomechanics, we propose tools based on the concept of Bregman divergence, a notion introduced in the 1960s and used in learning and optimization as well. This study is motivated by the need of “discrepancy measures” between physically constrained fields that are used both in traditional algorithms, analysis methods and data driven modelling or applications as well.We give also a characterization of symmetrical Bregman divergences through their generating functions which can only be quadratic forms. Some properties of the Bregman divergence and the introduced concept of Bregman Gap between couples of dual quantities are given, and some existing errors in thermomechanics are recovered. Exploiting the framework of Standard Generalized Materials, we give a set of generating functions for a large range of applications, including coupled multi-physics. Finally some results useful for computational geometry are detailed.https://comptes-rendus.academie-sciences.fr/mecanique/articles/10.5802/crmeca.164/ThermomechanicsConvexityBregman divergencesDiscrepancy measuresData processing
spellingShingle Andrieux, Stéphane
Bregman divergences for physically informed discrepancy measures for learning and computation in thermomechanics
Comptes Rendus. Mécanique
Thermomechanics
Convexity
Bregman divergences
Discrepancy measures
Data processing
title Bregman divergences for physically informed discrepancy measures for learning and computation in thermomechanics
title_full Bregman divergences for physically informed discrepancy measures for learning and computation in thermomechanics
title_fullStr Bregman divergences for physically informed discrepancy measures for learning and computation in thermomechanics
title_full_unstemmed Bregman divergences for physically informed discrepancy measures for learning and computation in thermomechanics
title_short Bregman divergences for physically informed discrepancy measures for learning and computation in thermomechanics
title_sort bregman divergences for physically informed discrepancy measures for learning and computation in thermomechanics
topic Thermomechanics
Convexity
Bregman divergences
Discrepancy measures
Data processing
url https://comptes-rendus.academie-sciences.fr/mecanique/articles/10.5802/crmeca.164/
work_keys_str_mv AT andrieuxstephane bregmandivergencesforphysicallyinformeddiscrepancymeasuresforlearningandcomputationinthermomechanics