Scaling Up Optuna: P2P Distributed Hyperparameters Optimization

ABSTRACT In machine learning (ML), hyperparameter optimization (HPO) is the process of choosing a tuple of values that ensures an efficient deployment and training of an AI model. In practice, HPO not only applies to ML tuning but can also be used to tune complex numerical simulations. In this conte...

Ausführliche Beschreibung

Gespeichert in:

Bibliographische Detailangaben
Veröffentlicht in:	Concurrency and computation Jg. 37; H. 4-5
1. Verfasser:	Cudennec, Loïc
Format:	Journal Article
Sprache:	Englisch
Veröffentlicht:	Hoboken, USA John Wiley & Sons, Inc 28.02.2025 Wiley Subscription Services, Inc Wiley
Schriftenreihe:	e70008
Schlagworte:	Algorithms Computer Science Computer simulation distributed computing Energy consumption high‐performance computing hyperparameters optimization Machine learning Numerical models operational research Optimization optuna Parameters peer‐to‐peer Scaling up Simulation Tuning Energy consumption Operational Research Peer-to-peer High-Performance Computing Optuna Distributed Computing Machine Learning Hyperparameters Optimization
ISSN:	1532-0626, 1532-0634
Online-Zugang:	Volltext
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!

Abstract	ABSTRACT In machine learning (ML), hyperparameter optimization (HPO) is the process of choosing a tuple of values that ensures an efficient deployment and training of an AI model. In practice, HPO not only applies to ML tuning but can also be used to tune complex numerical simulations. In this context, a numerical model of a given object is created to be used in realistic simulations. This model is defined by a set of values describing properties such as the geometry of the object or other unknown parameters related to physical quantities. While HPO for ML usually requires finding a few parameters, a numerical model can involve the tuning of more than a hundred parameters. As a consequence, a large number of tuples have to be explored and evaluated before finding a relevant solution, offering new challenges in high‐performance computing for efficiently driving the optimization. In this work we rely on the Optuna HPO framework, primarily designed for ML tasks and including state‐of‐the‐art sampling and pruning algorithms. We report on its use to optimize a complex numerical model onto a 1024‐core machine. We suggest 1.5M tuples and evaluate 5M simulations using different Optuna‐distributed layouts to build several tradeoffs between performance and energy consumption metrics. In order to further scale up the optimization process onto resources, we introduce OptunaP2P, an extension of Optuna based on the peer‐to‐peer paradigm. This allows to remove any bottleneck in the management of the shared knowledge between optimization processes. With OptunaP2P, we were able to compute up to 3 times faster compared to the regular Optuna‐distributed implementation and to obtain close‐to‐similar results in terms of quality in this reduced time‐frame.
AbstractList	ABSTRACT In machine learning (ML), hyperparameter optimization (HPO) is the process of choosing a tuple of values that ensures an efficient deployment and training of an AI model. In practice, HPO not only applies to ML tuning but can also be used to tune complex numerical simulations. In this context, a numerical model of a given object is created to be used in realistic simulations. This model is defined by a set of values describing properties such as the geometry of the object or other unknown parameters related to physical quantities. While HPO for ML usually requires finding a few parameters, a numerical model can involve the tuning of more than a hundred parameters. As a consequence, a large number of tuples have to be explored and evaluated before finding a relevant solution, offering new challenges in high‐performance computing for efficiently driving the optimization. In this work we rely on the Optuna HPO framework, primarily designed for ML tasks and including state‐of‐the‐art sampling and pruning algorithms. We report on its use to optimize a complex numerical model onto a 1024‐core machine. We suggest 1.5M tuples and evaluate 5M simulations using different Optuna‐distributed layouts to build several tradeoffs between performance and energy consumption metrics. In order to further scale up the optimization process onto resources, we introduce OptunaP2P, an extension of Optuna based on the peer‐to‐peer paradigm. This allows to remove any bottleneck in the management of the shared knowledge between optimization processes. With OptunaP2P, we were able to compute up to 3 times faster compared to the regular Optuna‐distributed implementation and to obtain close‐to‐similar results in terms of quality in this reduced time‐frame. In machine learning (ML), hyperparameter optimization (HPO) is the process of choosing a tuple of values that ensures an efficient deployment and training of an AI model. In practice, HPO not only applies to ML tuning but can also be used to tune complex numerical simulations. In this context, a numerical model of a given object is created to be used in realistic simulations. This model is defined by a set of values describing properties such as the geometry of the object or other unknown parameters related to physical quantities. While HPO for ML usually requires finding a few parameters, a numerical model can involve the tuning of more than a hundred parameters. As a consequence, a large number of tuples have to be explored and evaluated before finding a relevant solution, offering new challenges in high‐performance computing for efficiently driving the optimization. In this work we rely on the Optuna HPO framework, primarily designed for ML tasks and including state‐of‐the‐art sampling and pruning algorithms. We report on its use to optimize a complex numerical model onto a 1024‐core machine. We suggest 1.5M tuples and evaluate 5M simulations using different Optuna‐distributed layouts to build several tradeoffs between performance and energy consumption metrics. In order to further scale up the optimization process onto resources, we introduce OptunaP2P, an extension of Optuna based on the peer‐to‐peer paradigm. This allows to remove any bottleneck in the management of the shared knowledge between optimization processes. With OptunaP2P, we were able to compute up to 3 times faster compared to the regular Optuna‐distributed implementation and to obtain close‐to‐similar results in terms of quality in this reduced time‐frame.
Author	Cudennec, Loïc
Author_xml	– sequence: 1 givenname: Loïc surname: Cudennec fullname: Cudennec, Loïc email: loic.cudennec@intradef.gouv.fr organization: Ministry of Armed Forces
BackLink	https://hal.science/hal-05170088$$DView record in HAL
BookMark	eNp1kE9PwkAQxTcGEwE9-A2aePJQ2D8tnXoxBFFMSCBRzpuh3eqS0q67rQY_vS01ePI0LzO_eZl5A9IrykIRcs3oiFHKx4lRo4hSCmekz0LBfToRQe-k-eSCDJzbUcoYFaxP7l8SzHXx5m2MtzJVXeCdt-Zr70G7yuptXanUWxyMsgYt7lWlrGs5vdffWOmyuCTnGeZOXf3WIdk8zl9nC3-5enqeTZd-wjkDHyaBiEXCBcZxAjGqIOMxICjWTMMIeZDF0VaASPGogYcBoIggzNIMUhBDctv5vmMujdV7tAdZopaL6VK2PRqy5m-AT9awNx1rbPlRK1fJXVnbojlPChYFTEQ05H-OiS2dsyo72TIq2yxlk6U8Ztmw44790rk6_A_K2XrebfwAdUF0Rg
Cites_doi	10.25080/Majora-8b375195-004 10.1109/TC.2018.2883597 10.1177/1094342017727061 10.1145/1925861.1925881 10.1145/2939672.2939785 10.1002/CPE.5573 10.1109/TPDS.2004.10 10.1016/J.JPDC.2023.02.004 10.1007/S11227‐023‐05506‐7 10.1016/J.BSPC.2021.103456 10.1016/J.JPDC.2005.06.014 10.1145/3292500.3330701 10.1016/J.JAG.2023.103446 10.1016/J.FUTURE.2023.10.002 10.1109/IPDPSW.2013.142 10.1109/PDP.2010.67 10.1002/cpe.5960 10.1109/JSAC.2003.818784 10.1109/IPDPSW55747.2022.00173 10.1145/3555776.3577847
ContentType	Journal Article
Copyright	2025 John Wiley & Sons Ltd. Attribution - NonCommercial
Copyright_xml	– notice: 2025 John Wiley & Sons Ltd. – notice: Attribution - NonCommercial
DBID	AAYXX CITATION 7SC 8FD JQ2 L7M L~C L~D 1XC VOOES
DOI	10.1002/cpe.70008
DatabaseName	CrossRef Computer and Information Systems Abstracts Technology Research Database ProQuest Computer Science Collection Advanced Technologies Database with Aerospace Computer and Information Systems Abstracts Academic Computer and Information Systems Abstracts Professional Hyper Article en Ligne (HAL) Hyper Article en Ligne (HAL) (Open Access)
DatabaseTitle	CrossRef Computer and Information Systems Abstracts Technology Research Database Computer and Information Systems Abstracts – Academic Advanced Technologies Database with Aerospace ProQuest Computer Science Collection Computer and Information Systems Abstracts Professional
DatabaseTitleList	CrossRef Computer and Information Systems Abstracts
DeliveryMethod	fulltext_linktorsrc
Discipline	Computer Science
EISSN	1532-0634
EndPage	n/a
ExternalDocumentID	oai:HAL:hal-05170088v1 10_1002_cpe_70008 CPE70008
Genre	researchArticle
GroupedDBID	.3N .DC .GA 05W 0R~ 10A 1L6 1OC 33P 3SF 3WU 4.4 50Y 50Z 51W 51X 52M 52N 52O 52P 52S 52T 52U 52W 52X 5GY 5VS 66C 702 7PT 8-0 8-1 8-3 8-4 8-5 8UM 930 A03 AAESR AAEVG AAHHS AAHQN AAMNL AANLZ AAONW AAXRX AAYCA AAZKR ABCQN ABCUV ABEML ABIJN ACAHQ ACCFJ ACCZN ACPOU ACSCC ACXBN ACXQS ADBBV ADEOM ADIZJ ADKYN ADMGS ADOZA ADXAS ADZMN ADZOD AEEZP AEIGN AEIMD AEQDE AEUYR AFBPY AFFPM AFGKR AFWVQ AHBTC AITYG AIURR AIWBW AJBDE AJXKR ALMA_UNASSIGNED_HOLDINGS ALUQN ALVPJ AMBMR AMYDB ATUGU AUFTA AZBYB BAFTC BDRZF BFHJK BHBCM BMNLL BROTX BRXPI BY8 CS3 D-E D-F DCZOG DPXWK DR2 DRFUL DRSTM EBS F00 F01 F04 F5P G-S G.N GNP GODZA HGLYW HHY HZ~ IX1 JPC KQQ LATKE LAW LC2 LC3 LEEKS LITHE LOXES LP6 LP7 LUTES LYRES MEWTI MK4 MRFUL MRSTM MSFUL MSSTM MXFUL MXSTM N04 N05 N9A O66 O9- OIG P2W P2X P4D PQQKQ Q.N Q11 QB0 QRW R.K ROL RX1 SUPJJ TN5 UB1 V2E W8V W99 WBKPD WIH WIK WOHZO WQJ WXSBR WYISQ WZISG XG1 XV2 ~IA ~WT AAYXX AEYWJ AGHNM AGYGG CITATION LH4 O8X 1OB 7SC 8FD JQ2 L7M L~C L~D 1XC VOOES
ID	FETCH-LOGICAL-c2218-864393c23a99c89ae4f298a8e121857a24f97b383daa24f982548a3785fdf8d83
IEDL.DBID	DRFUL
ISICitedReferencesCount	0
ISICitedReferencesURI	http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=001436603800001&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
ISSN	1532-0626
IngestDate	Sun Oct 19 06:20:22 EDT 2025 Wed Aug 13 07:48:53 EDT 2025 Sat Nov 29 07:51:29 EST 2025 Thu Mar 06 09:30:41 EST 2025
IsDoiOpenAccess	true
IsOpenAccess	true
IsPeerReviewed	true
IsScholarly	true
Issue	4-5
Keywords	Energy consumption Operational Research Peer-to-peer High-Performance Computing Optuna Distributed Computing Machine Learning Hyperparameters Optimization
Language	English
License	Attribution - NonCommercial: http://creativecommons.org/licenses/by-nc
LinkModel	DirectLink
MergedId	FETCHMERGED-LOGICAL-c2218-864393c23a99c89ae4f298a8e121857a24f97b383daa24f982548a3785fdf8d83
Notes	ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 14
ORCID	0000-0002-6476-4574
OpenAccessLink	https://hal.science/hal-05170088
PQID	3174137052
PQPubID	2045170
PageCount	22
ParticipantIDs	hal_primary_oai_HAL_hal_05170088v1 proquest_journals_3174137052 crossref_primary_10_1002_cpe_70008 wiley_primary_10_1002_cpe_70008_CPE70008
PublicationCentury	2000
PublicationDate	28 February 2025
PublicationDateYYYYMMDD	2025-02-28
PublicationDate_xml	– month: 02 year: 2025 text: 28 February 2025 day: 28
PublicationDecade	2020
PublicationPlace	Hoboken, USA
PublicationPlace_xml	– name: Hoboken, USA – name: Hoboken
PublicationSeriesTitle	e70008
PublicationTitle	Concurrency and computation
PublicationYear	2025
Publisher	John Wiley & Sons, Inc Wiley Subscription Services, Inc Wiley
Publisher_xml	– name: John Wiley & Sons, Inc – name: Wiley Subscription Services, Inc – name: Wiley
References	2004; 22 2010 2022; 73 2024; 80 2023; 122 2020; 32 2023 2004; 34 2004 2001 2022 2013; 13 2021 2020 2006; 66 2019; 68 2004; 15 2023; 176 2024; 151 2019 2011; 41 2018 2016 2015 2013 2018; 32 Otaki R. (e_1_2_9_10_1) 2022 Foster I. T. (e_1_2_9_33_1) Zuber A. (e_1_2_9_6_1) 2022 Moritz P. (e_1_2_9_3_1) 2018 Stoica I. (e_1_2_9_31_1) 2001 Kennedy P. (e_1_2_9_34_1) 2019 e_1_2_9_35_1 e_1_2_9_13_1 e_1_2_9_32_1 e_1_2_9_12_1 Meng X. (e_1_2_9_9_1) 2015 Rosendo D. (e_1_2_9_36_1) 2020 Heckmann O. (e_1_2_9_30_1) 2004 e_1_2_9_15_1 e_1_2_9_14_1 e_1_2_9_17_1 Trabelsi K. (e_1_2_9_16_1) 2019 e_1_2_9_19_1 e_1_2_9_18_1 e_1_2_9_20_1 e_1_2_9_22_1 e_1_2_9_21_1 e_1_2_9_24_1 Contributors (e_1_2_9_4_1) 2018 e_1_2_9_23_1 e_1_2_9_8_1 e_1_2_9_7_1 e_1_2_9_2_1 e_1_2_9_26_1 Rozencwajg H. (e_1_2_9_11_1) 2021 e_1_2_9_25_1 e_1_2_9_28_1 e_1_2_9_27_1 e_1_2_9_29_1 Kozuki M. (e_1_2_9_5_1) 2020
References_xml	– volume: 41 start-page: 120 issue: 1 year: 2011 end-page: 124 article-title: Topology‐Aware Resource Allocation for Data‐Intensive Workloads publication-title: Proceedings of the First ACM Asia‐Pacific Workshop on Workshop on Systems – volume: 122 year: 2023 article-title: Quantifying Scattering Characteristics of Mangrove Species From Optuna‐Based Optimal Machine Learning Classification Using Multi‐Scale Feature Selection and SAR Image Time Series publication-title: International Journal of Applied Earth Observation and Geoinformation – volume: 32 start-page: 14 issue: 1 year: 2018 end-page: 27 article-title: Topology‐Aware Job Mapping publication-title: International Journal of High Performance Computing Applications – volume: 15 start-page: 546 issue: 6 year: 2004 end-page: 558 article-title: Mapping and Load‐Balancing Iterative Computations publication-title: IEEE Transactions on Parallel and Distributed Systems – volume: 22 start-page: 41 issue: 1 year: 2004 end-page: 53 article-title: Tapestry: A Resilient Global‐Scale Overlay for Service Deployment publication-title: IEEE Journal on Selected Areas in Communications – volume: 32 issue: 24 year: 2020 article-title: Adaptive Message Passing Polling for Energy Efficiency: Application to Software‐Distributed Shared Memory Over Heterogeneous Computing Resources publication-title: Concurrency and Computation: Practice and Experience – start-page: 19 year: 2013 end-page: 30 – volume: 66 start-page: 32 issue: 1 year: 2006 end-page: 46 article-title: Task Assignment in Heterogeneous Computing Systems publication-title: Journal of Parallel and Distributed Computing – start-page: 1053 year: 2022 end-page: 1062 – year: 2021 – start-page: 176 year: 2020 end-page: 186 – start-page: 1209 year: 2023 end-page: 1216 – start-page: 149 year: 2001 end-page: 160 – volume: 73 year: 2022 article-title: Hyoptxg: OPTUNA Hyper‐Parameter Optimization Framework for Predicting Cardiovascular Disease Using XGBoost publication-title: Biomedical Signal Processing and Control – volume: 34 2004 start-page: 224 year: 2004 end-page: 228 – volume: 176 start-page: 1 year: 2023 end-page: 16 article-title: Energy‐Aware Mapping and Scheduling Strategies for Real‐Time Workflows Under Reliability Constraints publication-title: Journal of Parallel and Distributed Computing – start-page: 561 year: 2018 end-page: 577 – year: 2018 – volume: 32 issue: 15 year: 2020 article-title: Efficient Algorithm for Scheduling Parallel Applications on Hybrid Multicore Machines With Communications Delays and Energy Constraint publication-title: Concurrency and Computation: Practice and Experience – start-page: 785 year: 2016 end-page: 794 – start-page: 118 end-page: 128 – year: 2015 article-title: MLlib: Machine Learning in Apache Spark publication-title: Journal of Machine Learning Research – year: 2022 – year: 2020 – start-page: 258 year: 2019 end-page: 269 – volume: 151 start-page: 214 year: 2024 end-page: 231 article-title: Energy Efficient Task Scheduling Based on Deep Reinforcement Learning in Cloud Environment: A Specialized Review publication-title: Future Generation Computer Systems – start-page: 180 year: 2010 end-page: 186 – volume: 13 start-page: 20 year: 2013 article-title: Hyperopt: A Python Library for Optimizing the Hyperparameters of Machine Learning Algorithms publication-title: SciPy – year: 2019 – start-page: 2623 year: 2019 end-page: 2631 – volume: 68 start-page: 713 issue: 5 year: 2019 end-page: 728 article-title: mARGOt: A Dynamic Autotuning Framework for Self‐Aware Approximate Computing publication-title: IEEE Transactions on Computers – volume: 80 start-page: 549 issue: 1 year: 2024 end-page: 569 article-title: Carbon Emission‐Aware Job Scheduling for Kubernetes Deployments publication-title: Journal of Supercomputing – ident: e_1_2_9_7_1 doi: 10.25080/Majora-8b375195-004 – ident: e_1_2_9_15_1 doi: 10.1109/TC.2018.2883597 – ident: e_1_2_9_13_1 doi: 10.1177/1094342017727061 – volume-title: Hyperparameter Optimization Run Time and Cost Using AWS and Optuna year: 2021 ident: e_1_2_9_11_1 – ident: e_1_2_9_14_1 doi: 10.1145/1925861.1925881 – ident: e_1_2_9_28_1 doi: 10.1145/2939672.2939785 – ident: e_1_2_9_21_1 doi: 10.1002/CPE.5573 – start-page: 118 volume-title: Peer‐To‐Peer Systems II, Second International Workshop, IPTPS 2003 ident: e_1_2_9_33_1 – start-page: 149 volume-title: Proceedings of the ACM SIGCOMM 2001 Conference on Applications, Technologies, Architectures, and Protocols for Computer Communication year: 2001 ident: e_1_2_9_31_1 – ident: e_1_2_9_19_1 doi: 10.1109/TPDS.2004.10 – ident: e_1_2_9_20_1 doi: 10.1016/J.JPDC.2023.02.004 – ident: e_1_2_9_23_1 doi: 10.1007/S11227‐023‐05506‐7 – ident: e_1_2_9_27_1 doi: 10.1016/J.BSPC.2021.103456 – ident: e_1_2_9_18_1 doi: 10.1016/J.JPDC.2005.06.014 – ident: e_1_2_9_2_1 doi: 10.1145/3292500.3330701 – volume-title: AMD EPYC 7742 Benchmarks and Review Simply Peerless year: 2019 ident: e_1_2_9_34_1 – start-page: 258 volume-title: Parallel Processing Workshops–Euro‐Par 2019 International Workshops 11997 of Lecture Notes in Computer Science year: 2019 ident: e_1_2_9_16_1 – start-page: 224 volume-title: Jahrestagung der Gesellschaft für Informatik, Informatik verbindet, INFORMATIK 2004 year: 2004 ident: e_1_2_9_30_1 – ident: e_1_2_9_29_1 doi: 10.1016/J.JAG.2023.103446 – volume-title: Parallel Hyperparameter Tuning With Optuna and Kubeflow Pipelines year: 2020 ident: e_1_2_9_5_1 – volume-title: Distributed Optimization via NFS Using Optuna's New Operation‐Based Logging Storage year: 2022 ident: e_1_2_9_10_1 – ident: e_1_2_9_24_1 doi: 10.1016/J.FUTURE.2023.10.002 – volume-title: Optuna–Easy Parallelization year: 2018 ident: e_1_2_9_4_1 – ident: e_1_2_9_17_1 doi: 10.1109/IPDPSW.2013.142 – start-page: 176 volume-title: E2Clab: Exploring the Computing Continuum Through Repeatable, Replicable and Reproducible Edge‐To‐Cloud Experiments year: 2020 ident: e_1_2_9_36_1 – year: 2015 ident: e_1_2_9_9_1 article-title: MLlib: Machine Learning in Apache Spark publication-title: Journal of Machine Learning Research – ident: e_1_2_9_12_1 doi: 10.1109/PDP.2010.67 – ident: e_1_2_9_35_1 doi: 10.1002/cpe.5960 – ident: e_1_2_9_32_1 doi: 10.1109/JSAC.2003.818784 – ident: e_1_2_9_8_1 – volume-title: Running Distributed Hyperparameter Optimization With Optuna‐Distributed year: 2022 ident: e_1_2_9_6_1 – start-page: 561 volume-title: 13th USENIX Symposium on Operating Systems Design and Implementation year: 2018 ident: e_1_2_9_3_1 – ident: e_1_2_9_26_1 – ident: e_1_2_9_22_1 doi: 10.1109/IPDPSW55747.2022.00173 – ident: e_1_2_9_25_1 doi: 10.1145/3555776.3577847
SSID	ssj0011031
Score	2.3886168
Snippet	ABSTRACT In machine learning (ML), hyperparameter optimization (HPO) is the process of choosing a tuple of values that ensures an efficient deployment and... In machine learning (ML), hyperparameter optimization (HPO) is the process of choosing a tuple of values that ensures an efficient deployment and training of...
SourceID	hal proquest crossref wiley
SourceType	Open Access Repository Aggregation Database Index Database Publisher
SubjectTerms	Algorithms Computer Science Computer simulation distributed computing Energy consumption high‐performance computing hyperparameters optimization Machine learning Numerical models operational research Optimization optuna Parameters peer‐to‐peer Scaling up Simulation Tuning
Title	Scaling Up Optuna: P2P Distributed Hyperparameters Optimization
URI	https://onlinelibrary.wiley.com/doi/abs/10.1002%2Fcpe.70008 https://www.proquest.com/docview/3174137052 https://hal.science/hal-05170088
Volume	37
WOSCitedRecordID	wos001436603800001&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
hasFullText	1
inHoldings	1
isFullTextHit
isPrint
journalDatabaseRights	– providerCode: PRVWIB databaseName: Wiley Online Library Full Collection 2020 customDbUrl: eissn: 1532-0634 dateEnd: 99991231 omitProxy: false ssIdentifier: ssj0011031 issn: 1532-0626 databaseCode: DRFUL dateStart: 20010101 isFulltext: true titleUrlDefault: https://onlinelibrary.wiley.com providerName: Wiley-Blackwell
link	http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwpV3NT8IwFG8QPHgRPyOKpjEevExGO2irJ8JHOBBcjCTcmrJ1kQNzYcDfb1_3IR5MTLw121u2vPbt_d7W9_sh9KDDhSBdSh2T7QLHU4Q5iobEMVCBi6BLFkxZyvwJm075fC78CnopemEyfojygxtEhn1fQ4CrRdr6Jg0NEv3EXNvoWyNm3XpVVBu8jWaT8icCKBhkdKnEcQ1wL4iFXNIqL_6Rjg4-YDPkHtLcx6s24Yzq_3rUE3Sc40zcyxbGKaro-AzVCw0HnIf0uQHrZpJM-sKzBL8mm22snrFPfDwAQl3QwtIhHptadQ0c4SvYO5OC3XKV929eoNlo-N4fO7moghMQk84dDhCEBoQqIQIulPYiIrjiuk2AFkoRLxJsYerWUNkxVJBcUcY7URjxkNNLVI0_Y32FcBh5HcYirTkwxgjFoZhxQVe8LXRAVQPdF76VScadITOWZCKNS6R1iTEyXi_PA9v1uDeRcAzow4wF37UbqFlMisxjLJUG-ZgMzNwOaaBH6_7f7yL7_tAOrv9ueoOOCIj92v71Jqpu1lt9iw6D3WaZru_yxfYFxDzTaA
linkProvider	Wiley-Blackwell
linkToHtml	http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwpV1LT8JAEJ4gmOhFfEYUtTEevFTKtrC7xgvhEYwViYGE22Zpt5EDlfD6_e5sW8SDiYm3TTtNm9mZzje7O98A3KlwzEnddW0d7QLbk4Ta0g2JraEC40GdjKk0lPk-7fXYaMT7OXjKamESfojNght6hvlfo4PjgnTlmzU0mKkH6phK34KnzaiWh0LrvTP0N7sI2MIg4UsltqORe8Ys5JDK5uEf8WjnA09DbkHNbcBqIk6n-L9vPYSDFGlajcQ0jiCn4mMoZl0crNSpTzRc19OkA5g1nFlvs-Uqlo9Wn_StFlLqYjcsFVpdna3OkSV8iqdnFig3maYVnKcw7LQHza6dtlWwA6IDus0QhLgBcSXnAeNSeRHhTDJVJUgMJYkXcTrWmWsozRhzSCZdympRGLGQuWeQjz9jdQ5WGHk1SiOlGHLGcMkwnXGws3iVq8CVJbjNlCtmCXuGSHiSidAqEUYlWkirfXMf-a67DV_gNSQQ0xJsXS1BOZsVkXrZQmjso2MwdWqkBPdG_7-_RTT7bTO4-LvoDex1B6--8J97L5ewT7D1r6lmL0N-OV-pK9gN1svJYn6dWt4Xf9jXWA
linkToPdf	http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwpV1LS8NAEF5qK-LF-sRq1SAevMSmm6S7K15K21Ax1CAWelu2yQZ7aAx9_X53Ng_rQRC8LcmEhNmdfN8kO98gdCejKcMd2zYV2oWmIzAxhR1hU1EFysIOnhKhJfN9MhrRyYQFFfRU1MJk-hDlBzeIDP2-hgCXaRS3vlVDw1Q-EEtX-tYcl3VUWNb6b97YL_8iQAuDTC8Vm5Zi7oWykIVb5cU_8GjnA3ZDblHNbcKqEcer_-9ZD9FBzjSNbrY0jlBFJseoXnRxMPKgPlF0XU2TAjBjnBqv6WqdiEcjwIHRB0ld6IYlI2OostUFqITPYffMEuxm87yC8xSNvcF7b2jmbRXMECtANymQEDvEtmAspExIJ8aMCirbGIShBHZiRqYqc42EHkMOSYVNqBtHMY2ofYaqyWciz5ERxY5LSCwlBc0YJiikMxZ0Fm8zGdqigW4L5_I0U8_gmU4y5solXLtEGSm3l-dB73rY9TkcAwExZUE37QZqFrPC8yhbcsV9FAYTy8UNdK_9__tdeC8Y6MHF301v0F7Q97j_PHq5RPsYOv_qYvYmqq4Wa3mFdsPNarZcXOcL7wtFQNbT
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=Scaling+Up+Optuna%3A+P2P+Distributed+Hyperparameters+Optimization&rft.jtitle=Concurrency+and+computation&rft.au=Cudennec%2C+Lo%C3%AFc&rft.date=2025-02-28&rft.pub=Wiley+Subscription+Services%2C+Inc&rft.issn=1532-0626&rft.eissn=1532-0634&rft.volume=37&rft.issue=4-5&rft_id=info:doi/10.1002%2Fcpe.70008&rft.externalDBID=NO_FULL_TEXT
thumbnail_l	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=1532-0626&client=summon
thumbnail_m	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=1532-0626&client=summon
thumbnail_s	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=1532-0626&client=summon