Dropout Ensemble Kalman inversion for high dimensional inverse problems

Ensemble Kalman inversion (EKI) is an ensemble-based method to solve inverse problems. Its gradient-free formulation makes it an attractive tool for problems with involved formulation. However, EKI suffers from the ''subspace property'', i.e., the EKI solutions are confined in th...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:arXiv.org 2023-08
Hauptverfasser: Liu, Shuigen, Reich, Sebastian, Tong, Xin T
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
container_end_page
container_issue
container_start_page
container_title arXiv.org
container_volume
creator Liu, Shuigen
Reich, Sebastian
Tong, Xin T
description Ensemble Kalman inversion (EKI) is an ensemble-based method to solve inverse problems. Its gradient-free formulation makes it an attractive tool for problems with involved formulation. However, EKI suffers from the ''subspace property'', i.e., the EKI solutions are confined in the subspace spanned by the initial ensemble. It implies that the ensemble size should be larger than the problem dimension to ensure EKI's convergence to the correct solution. Such scaling of ensemble size is impractical and prevents the use of EKI in high dimensional problems. To address this issue, we propose a novel approach using dropout regularization to mitigate the subspace problem. We prove that dropout-EKI converges in the small ensemble settings, and the computational cost of the algorithm scales linearly with dimension. We also show that dropout-EKI reaches the optimal query complexity, up to a constant factor. Numerical examples demonstrate the effectiveness of our approach.
format Article
fullrecord <record><control><sourceid>proquest</sourceid><recordid>TN_cdi_proquest_journals_2859741784</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2859741784</sourcerecordid><originalsourceid>FETCH-proquest_journals_28597417843</originalsourceid><addsrcrecordid>eNqNi80KgkAURocgSMp3uNBa0JkxbV1W0La9jHTNkfmxudrzZ-ADtPrgfOesWMSFyJJScr5hMVGfpik_FDzPRcSu5-AHP41QOULbGIS7MlY50O6DgbR30PoAnX518NQW3Q8ps9wIQ_BzZGnH1q0yhPGyW7a_VI_TLZmF94Q01r2fwlxSzcv8WMisKKX4z_oCR408XQ</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2859741784</pqid></control><display><type>article</type><title>Dropout Ensemble Kalman inversion for high dimensional inverse problems</title><source>Freely Accessible Journals</source><creator>Liu, Shuigen ; Reich, Sebastian ; Tong, Xin T</creator><creatorcontrib>Liu, Shuigen ; Reich, Sebastian ; Tong, Xin T</creatorcontrib><description>Ensemble Kalman inversion (EKI) is an ensemble-based method to solve inverse problems. Its gradient-free formulation makes it an attractive tool for problems with involved formulation. However, EKI suffers from the ''subspace property'', i.e., the EKI solutions are confined in the subspace spanned by the initial ensemble. It implies that the ensemble size should be larger than the problem dimension to ensure EKI's convergence to the correct solution. Such scaling of ensemble size is impractical and prevents the use of EKI in high dimensional problems. To address this issue, we propose a novel approach using dropout regularization to mitigate the subspace problem. We prove that dropout-EKI converges in the small ensemble settings, and the computational cost of the algorithm scales linearly with dimension. We also show that dropout-EKI reaches the optimal query complexity, up to a constant factor. Numerical examples demonstrate the effectiveness of our approach.</description><identifier>EISSN: 2331-8422</identifier><language>eng</language><publisher>Ithaca: Cornell University Library, arXiv.org</publisher><subject>Algorithms ; Convergence ; Inverse problems ; Regularization ; Subspaces</subject><ispartof>arXiv.org, 2023-08</ispartof><rights>2023. This work is published under http://creativecommons.org/licenses/by/4.0/ (the “License”). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.</rights><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>776,780</link.rule.ids></links><search><creatorcontrib>Liu, Shuigen</creatorcontrib><creatorcontrib>Reich, Sebastian</creatorcontrib><creatorcontrib>Tong, Xin T</creatorcontrib><title>Dropout Ensemble Kalman inversion for high dimensional inverse problems</title><title>arXiv.org</title><description>Ensemble Kalman inversion (EKI) is an ensemble-based method to solve inverse problems. Its gradient-free formulation makes it an attractive tool for problems with involved formulation. However, EKI suffers from the ''subspace property'', i.e., the EKI solutions are confined in the subspace spanned by the initial ensemble. It implies that the ensemble size should be larger than the problem dimension to ensure EKI's convergence to the correct solution. Such scaling of ensemble size is impractical and prevents the use of EKI in high dimensional problems. To address this issue, we propose a novel approach using dropout regularization to mitigate the subspace problem. We prove that dropout-EKI converges in the small ensemble settings, and the computational cost of the algorithm scales linearly with dimension. We also show that dropout-EKI reaches the optimal query complexity, up to a constant factor. Numerical examples demonstrate the effectiveness of our approach.</description><subject>Algorithms</subject><subject>Convergence</subject><subject>Inverse problems</subject><subject>Regularization</subject><subject>Subspaces</subject><issn>2331-8422</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2023</creationdate><recordtype>article</recordtype><sourceid>BENPR</sourceid><recordid>eNqNi80KgkAURocgSMp3uNBa0JkxbV1W0La9jHTNkfmxudrzZ-ADtPrgfOesWMSFyJJScr5hMVGfpik_FDzPRcSu5-AHP41QOULbGIS7MlY50O6DgbR30PoAnX518NQW3Q8ps9wIQ_BzZGnH1q0yhPGyW7a_VI_TLZmF94Q01r2fwlxSzcv8WMisKKX4z_oCR408XQ</recordid><startdate>20230831</startdate><enddate>20230831</enddate><creator>Liu, Shuigen</creator><creator>Reich, Sebastian</creator><creator>Tong, Xin T</creator><general>Cornell University Library, arXiv.org</general><scope>8FE</scope><scope>8FG</scope><scope>ABJCF</scope><scope>ABUWG</scope><scope>AFKRA</scope><scope>AZQEC</scope><scope>BENPR</scope><scope>BGLVJ</scope><scope>CCPQU</scope><scope>DWQXO</scope><scope>HCIFZ</scope><scope>L6V</scope><scope>M7S</scope><scope>PIMPY</scope><scope>PQEST</scope><scope>PQQKQ</scope><scope>PQUKI</scope><scope>PTHSS</scope></search><sort><creationdate>20230831</creationdate><title>Dropout Ensemble Kalman inversion for high dimensional inverse problems</title><author>Liu, Shuigen ; Reich, Sebastian ; Tong, Xin T</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-proquest_journals_28597417843</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2023</creationdate><topic>Algorithms</topic><topic>Convergence</topic><topic>Inverse problems</topic><topic>Regularization</topic><topic>Subspaces</topic><toplevel>online_resources</toplevel><creatorcontrib>Liu, Shuigen</creatorcontrib><creatorcontrib>Reich, Sebastian</creatorcontrib><creatorcontrib>Tong, Xin T</creatorcontrib><collection>ProQuest SciTech Collection</collection><collection>ProQuest Technology Collection</collection><collection>Materials Science &amp; Engineering Collection</collection><collection>ProQuest Central (Alumni Edition)</collection><collection>ProQuest Central UK/Ireland</collection><collection>ProQuest Central Essentials</collection><collection>ProQuest Central</collection><collection>Technology Collection (ProQuest)</collection><collection>ProQuest One Community College</collection><collection>ProQuest Central Korea</collection><collection>SciTech Premium Collection</collection><collection>ProQuest Engineering Collection</collection><collection>Engineering Database</collection><collection>Publicly Available Content Database</collection><collection>ProQuest One Academic Eastern Edition (DO NOT USE)</collection><collection>ProQuest One Academic</collection><collection>ProQuest One Academic UKI Edition</collection><collection>Engineering Collection</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Liu, Shuigen</au><au>Reich, Sebastian</au><au>Tong, Xin T</au><format>book</format><genre>document</genre><ristype>GEN</ristype><atitle>Dropout Ensemble Kalman inversion for high dimensional inverse problems</atitle><jtitle>arXiv.org</jtitle><date>2023-08-31</date><risdate>2023</risdate><eissn>2331-8422</eissn><abstract>Ensemble Kalman inversion (EKI) is an ensemble-based method to solve inverse problems. Its gradient-free formulation makes it an attractive tool for problems with involved formulation. However, EKI suffers from the ''subspace property'', i.e., the EKI solutions are confined in the subspace spanned by the initial ensemble. It implies that the ensemble size should be larger than the problem dimension to ensure EKI's convergence to the correct solution. Such scaling of ensemble size is impractical and prevents the use of EKI in high dimensional problems. To address this issue, we propose a novel approach using dropout regularization to mitigate the subspace problem. We prove that dropout-EKI converges in the small ensemble settings, and the computational cost of the algorithm scales linearly with dimension. We also show that dropout-EKI reaches the optimal query complexity, up to a constant factor. Numerical examples demonstrate the effectiveness of our approach.</abstract><cop>Ithaca</cop><pub>Cornell University Library, arXiv.org</pub><oa>free_for_read</oa></addata></record>
fulltext fulltext
identifier EISSN: 2331-8422
ispartof arXiv.org, 2023-08
issn 2331-8422
language eng
recordid cdi_proquest_journals_2859741784
source Freely Accessible Journals
subjects Algorithms
Convergence
Inverse problems
Regularization
Subspaces
title Dropout Ensemble Kalman inversion for high dimensional inverse problems
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-02-06T09%3A37%3A33IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest&rft_val_fmt=info:ofi/fmt:kev:mtx:book&rft.genre=document&rft.atitle=Dropout%20Ensemble%20Kalman%20inversion%20for%20high%20dimensional%20inverse%20problems&rft.jtitle=arXiv.org&rft.au=Liu,%20Shuigen&rft.date=2023-08-31&rft.eissn=2331-8422&rft_id=info:doi/&rft_dat=%3Cproquest%3E2859741784%3C/proquest%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_pqid=2859741784&rft_id=info:pmid/&rfr_iscdi=true