A descent algorithm for the optimal control of ReLU neural network informed PDEs based on approximate directional derivatives

We propose and analyze a numerical algorithm for solving a class of optimal control problems for learning-informed semilinear partial differential equations. The latter is a class of PDEs with constituents that are in principle unknown and are approximated by nonsmooth ReLU neural networks. We first...

Full description

Saved in:
Bibliographic Details
Main Authors Dong, Guozhi, Hintermüller, Michael, Papafitsoros, Kostas
Format Journal Article
LanguageEnglish
Published 14.10.2022
Subjects
Online AccessGet full text
DOI10.48550/arxiv.2210.07900

Cover

Abstract We propose and analyze a numerical algorithm for solving a class of optimal control problems for learning-informed semilinear partial differential equations. The latter is a class of PDEs with constituents that are in principle unknown and are approximated by nonsmooth ReLU neural networks. We first show that a direct smoothing of the ReLU network with the aim to make use of classical numerical solvers can have certain disadvantages, namely potentially introducing multiple solutions for the corresponding state equation. This motivates us to devise a numerical algorithm that treats directly the nonsmooth optimal control problem, by employing a descent algorithm inspired by a bundle-free method. Several numerical examples are provided and the efficiency of the algorithm is shown.
AbstractList We propose and analyze a numerical algorithm for solving a class of optimal control problems for learning-informed semilinear partial differential equations. The latter is a class of PDEs with constituents that are in principle unknown and are approximated by nonsmooth ReLU neural networks. We first show that a direct smoothing of the ReLU network with the aim to make use of classical numerical solvers can have certain disadvantages, namely potentially introducing multiple solutions for the corresponding state equation. This motivates us to devise a numerical algorithm that treats directly the nonsmooth optimal control problem, by employing a descent algorithm inspired by a bundle-free method. Several numerical examples are provided and the efficiency of the algorithm is shown.
Author Papafitsoros, Kostas
Dong, Guozhi
Hintermüller, Michael
Author_xml – sequence: 1
  givenname: Guozhi
  surname: Dong
  fullname: Dong, Guozhi
– sequence: 2
  givenname: Michael
  surname: Hintermüller
  fullname: Hintermüller, Michael
– sequence: 3
  givenname: Kostas
  surname: Papafitsoros
  fullname: Papafitsoros, Kostas
BackLink https://doi.org/10.48550/arXiv.2210.07900$$DView paper in arXiv
BookMark eNqFjjFuwkAQRbdIChJyACrmAjiLEwSUUTBKkQIhqK2NPYYV9ow1OzFQcPcsiD7VH33N-3pP5oGY0JjB2Cbvs8nEvjo5-S5J01jY6dzanrl8QImhQFJw9Y7F676BigV0j8Ct-sbVUDCpcA1cwRq_t0D4K7Em1CPLATxFoMESVosswI8L8WQC17bCpzigCKUXLNQzRaxE8Z1T32Hom8fK1QFf7vlshsts8_k1uonmrURazvlVOL8Jv_3_8QcMjk9T
ContentType Journal Article
Copyright http://arxiv.org/licenses/nonexclusive-distrib/1.0
Copyright_xml – notice: http://arxiv.org/licenses/nonexclusive-distrib/1.0
DBID AKZ
GOX
DOI 10.48550/arxiv.2210.07900
DatabaseName arXiv Mathematics
arXiv.org
DatabaseTitleList
Database_xml – sequence: 1
  dbid: GOX
  name: arXiv.org
  url: http://arxiv.org/find
  sourceTypes: Open Access Repository
DeliveryMethod fulltext_linktorsrc
ExternalDocumentID 2210_07900
GroupedDBID AKZ
GOX
ID FETCH-arxiv_primary_2210_079003
IEDL.DBID GOX
IngestDate Tue Jul 22 23:40:10 EDT 2025
IsDoiOpenAccess true
IsOpenAccess true
IsPeerReviewed false
IsScholarly false
Language English
LinkModel DirectLink
MergedId FETCHMERGED-arxiv_primary_2210_079003
OpenAccessLink https://arxiv.org/abs/2210.07900
ParticipantIDs arxiv_primary_2210_07900
PublicationCentury 2000
PublicationDate 2022-10-14
PublicationDateYYYYMMDD 2022-10-14
PublicationDate_xml – month: 10
  year: 2022
  text: 2022-10-14
  day: 14
PublicationDecade 2020
PublicationYear 2022
Score 3.6254845
SecondaryResourceType preprint
Snippet We propose and analyze a numerical algorithm for solving a class of optimal control problems for learning-informed semilinear partial differential equations....
SourceID arxiv
SourceType Open Access Repository
SubjectTerms Mathematics - Optimization and Control
Title A descent algorithm for the optimal control of ReLU neural network informed PDEs based on approximate directional derivatives
URI https://arxiv.org/abs/2210.07900
hasFullText 1
inHoldings 1
isFullTextHit
isPrint
link http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwdZ3PT4MwFMdf5k5ejEbN_P0OXolQYMBx0S2L8VeMS7iRQh9q4sAALrv4v_vaYvSya2lLf4S8b-m3nwJcJkoUmqPmyLDInUCo0pEq8J2oLFhce_k4MtyC-4fxfBHcpmE6APw9CyOb9fvK8oHz9koI7byKEpcX5VssFPRh3sfUbk4aFFef_y8fa0yT9C9IzHZhp1d3OLHTsQcDqvbhe4LKQpNQfrzWvBp_WyJrRWTthTV_sksu0lvGsS7xme4WqDmTnFxZlzZavikpfLqZtqhDj8K6QkMEX3MFHaENTubPHr-vMbeWrag9gIvZ9OV67pgGZ5-WLpHpvmSmL_4hDKu6ohGgLygml3zyZRl4Ms8TSmIvilTsuYV0wyMYbarlePOjE9gW2s2vDRrBKQy75ovOOMZ2-bkZ6B8bjINI
linkProvider Cornell University
openUrl ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=A+descent+algorithm+for+the+optimal+control+of+ReLU+neural+network+informed+PDEs+based+on+approximate+directional+derivatives&rft.au=Dong%2C+Guozhi&rft.au=Hinterm%C3%BCller%2C+Michael&rft.au=Papafitsoros%2C+Kostas&rft.date=2022-10-14&rft_id=info:doi/10.48550%2Farxiv.2210.07900&rft.externalDocID=2210_07900