FedPop: Federated Population-based Hyperparameter Tuning Article Swipe

PDF

Haokun Chen , Denis Krompaß , Jindong Gu , Volker Tresp ·

YOU? · · 2025 · Open Access · · DOI: https://doi.org/10.1609/aaai.v39i15.33732

Federated Learning (FL) is a distributed machine learning (ML) paradigm, in which multiple clients collaboratively train ML models without centralizing their local data. Similar to conventional ML pipelines, the client local optimization and server aggregation procedure in FL are sensitive to the hyperparameter (HP) selection. Despite extensive research on tuning HPs for centralized ML, these methods yield suboptimal results when employed in FL. This is mainly because their "training-after-tuning" framework is unsuitable for FL with limited client computation power. While some approaches have been proposed for HP-Tuning in FL, they are limited to the HPs for client local updates. In this work, we propose a novel HP-tuning algorithm, called Federated Population-based Hyperparameter Tuning (FedPop), to address this vital yet challenging problem. FedPop employs population-based evolutionary algorithms to optimize the HPs, which accommodates various HP types at both the client and server sides. Compared with prior tuning methods, FedPop employs an online "tuning-while-training" framework, offering computational efficiency and enabling the exploration of a broader HP search space. Our empirical validation on the common FL benchmarks and complex real-world FL datasets, including full-sized Non-IID ImageNet-1K, demonstrates the effectiveness of the proposed method, which substantially outperforms the concurrent state-of-the-art HP-tuning methods in FL.

Related Topics

Computer Science

Artificial Intelligence

Demography

Concepts

Hyperparameter Population Computer science Artificial intelligence Demography Sociology

Metadata

Type: article
Language: en
Landing Page: https://doi.org/10.1609/aaai.v39i15.33732
PDF: https://ojs.aaai.org/index.php/AAAI/article/download/33732/35887
OA Status: diamond
Cited By: 1
Related Works: 10
OpenAlex ID: https://openalex.org/W4409347486

All OpenAlex metadata

Raw OpenAlex JSON

OpenAlex ID: https://openalex.org/W4409347486

Canonical identifier for this work in OpenAlex
DOI: https://doi.org/10.1609/aaai.v39i15.33732

Digital Object Identifier
Title: FedPop: Federated Population-based Hyperparameter Tuning

Work title
Type: article

OpenAlex work type
Language: en

Primary language
Publication year: 2025

Year of publication
Publication date: 2025-04-11

Full publication date if available
Authors: Haokun Chen, Denis Krompaß, Jindong Gu, Volker Tresp

List of authors in order
Landing page: https://doi.org/10.1609/aaai.v39i15.33732

Publisher landing page
PDF URL: https://ojs.aaai.org/index.php/AAAI/article/download/33732/35887

Direct link to full text PDF
Open access: Yes

Whether a free full text is available
OA status: diamond

Open access status per OpenAlex
OA URL: https://ojs.aaai.org/index.php/AAAI/article/download/33732/35887

Direct OA link when available
Concepts: Hyperparameter, Population, Computer science, Artificial intelligence, Demography, Sociology

Top concepts (fields/topics) attached by OpenAlex
Cited by: 1

Total citation count in OpenAlex
Citations by year (recent): 2025: 1

Per-year citation counts (last 5 years)
Related works (count): 10

Other works algorithmically related by OpenAlex

Full payload

id	https://openalex.org/W4409347486
doi	https://doi.org/10.1609/aaai.v39i15.33732
ids.doi	https://doi.org/10.1609/aaai.v39i15.33732
ids.openalex	https://openalex.org/W4409347486
fwci	1.8235294
type	article
title	FedPop: Federated Population-based Hyperparameter Tuning
biblio.issue	15
biblio.volume	39
biblio.last_page	15784
biblio.first_page	15776
topics[0].id	https://openalex.org/T12205
topics[0].field.id	https://openalex.org/fields/17
topics[0].field.display_name	Computer Science
topics[0].score	0.9332000017166138
topics[0].domain.id	https://openalex.org/domains/3
topics[0].domain.display_name	Physical Sciences
topics[0].subfield.id	https://openalex.org/subfields/1711
topics[0].subfield.display_name	Signal Processing
topics[0].display_name	Time Series Analysis and Forecasting
is_xpac	False
apc_list
apc_paid
concepts[0].id	https://openalex.org/C8642999
concepts[0].level	2
concepts[0].score	0.8816965222358704
concepts[0].wikidata	https://www.wikidata.org/wiki/Q4171168
concepts[0].display_name	Hyperparameter
concepts[1].id	https://openalex.org/C2908647359
concepts[1].level	2
concepts[1].score	0.4854162633419037
concepts[1].wikidata	https://www.wikidata.org/wiki/Q2625603
concepts[1].display_name	Population
concepts[2].id	https://openalex.org/C41008148
concepts[2].level	0
concepts[2].score	0.39365291595458984
concepts[2].wikidata	https://www.wikidata.org/wiki/Q21198
concepts[2].display_name	Computer science
concepts[3].id	https://openalex.org/C154945302
concepts[3].level	1
concepts[3].score	0.19218644499778748
concepts[3].wikidata	https://www.wikidata.org/wiki/Q11660
concepts[3].display_name	Artificial intelligence
concepts[4].id	https://openalex.org/C149923435
concepts[4].level	1
concepts[4].score	0.1086851954460144
concepts[4].wikidata	https://www.wikidata.org/wiki/Q37732
concepts[4].display_name	Demography
concepts[5].id	https://openalex.org/C144024400
concepts[5].level	0
concepts[5].score	0.04597058892250061
concepts[5].wikidata	https://www.wikidata.org/wiki/Q21201
concepts[5].display_name	Sociology
keywords[0].id	https://openalex.org/keywords/hyperparameter
keywords[0].score	0.8816965222358704
keywords[0].display_name	Hyperparameter
keywords[1].id	https://openalex.org/keywords/population
keywords[1].score	0.4854162633419037
keywords[1].display_name	Population
keywords[2].id	https://openalex.org/keywords/computer-science
keywords[2].score	0.39365291595458984
keywords[2].display_name	Computer science
keywords[3].id	https://openalex.org/keywords/artificial-intelligence
keywords[3].score	0.19218644499778748
keywords[3].display_name	Artificial intelligence
keywords[4].id	https://openalex.org/keywords/demography
keywords[4].score	0.1086851954460144
keywords[4].display_name	Demography
keywords[5].id	https://openalex.org/keywords/sociology
keywords[5].score	0.04597058892250061
keywords[5].display_name	Sociology
language	en
locations[0].id	doi:10.1609/aaai.v39i15.33732
locations[0].is_oa	True
locations[0].source.id	https://openalex.org/S4210191458
locations[0].source.issn	2159-5399, 2374-3468
locations[0].source.type	conference
locations[0].source.is_oa	True
locations[0].source.issn_l	2159-5399
locations[0].source.is_core	False
locations[0].source.is_in_doaj	False
locations[0].source.display_name	Proceedings of the AAAI Conference on Artificial Intelligence
locations[0].source.host_organization	https://openalex.org/P4310320058
locations[0].source.host_organization_name	Association for the Advancement of Artificial Intelligence
locations[0].source.host_organization_lineage	https://openalex.org/P4310320058
locations[0].source.host_organization_lineage_names	Association for the Advancement of Artificial Intelligence
locations[0].license
locations[0].pdf_url	https://ojs.aaai.org/index.php/AAAI/article/download/33732/35887
locations[0].version	publishedVersion
locations[0].raw_type	journal-article
locations[0].license_id
locations[0].is_accepted	True
locations[0].is_published	True
locations[0].raw_source_name	Proceedings of the AAAI Conference on Artificial Intelligence
locations[0].landing_page_url	https://doi.org/10.1609/aaai.v39i15.33732
indexed_in	crossref
authorships[0].author.id	https://openalex.org/A5045308982
authorships[0].author.orcid	https://orcid.org/0000-0002-5485-2984
authorships[0].author.display_name	Haokun Chen
authorships[0].author_position	first
authorships[0].raw_author_name	Haokun Chen
authorships[0].is_corresponding	False
authorships[1].author.id	https://openalex.org/A5055134806
authorships[1].author.orcid
authorships[1].author.display_name	Denis Krompaß
authorships[1].author_position	middle
authorships[1].raw_author_name	Denis Krompaß
authorships[1].is_corresponding	False
authorships[2].author.id	https://openalex.org/A5055994909
authorships[2].author.orcid	https://orcid.org/0009-0000-0574-0129
authorships[2].author.display_name	Jindong Gu
authorships[2].author_position	middle
authorships[2].raw_author_name	Jindong Gu
authorships[2].is_corresponding	False
authorships[3].author.id	https://openalex.org/A5074808403
authorships[3].author.orcid	https://orcid.org/0000-0001-9428-3686
authorships[3].author.display_name	Volker Tresp
authorships[3].author_position	last
authorships[3].raw_author_name	Volker Tresp
authorships[3].is_corresponding	False
has_content.pdf	True
has_content.grobid_xml	True
is_paratext	False
open_access.is_oa	True
open_access.oa_url	https://ojs.aaai.org/index.php/AAAI/article/download/33732/35887
open_access.oa_status	diamond
open_access.any_repository_has_fulltext	False
created_date	2025-10-10T00:00:00
display_name	FedPop: Federated Population-based Hyperparameter Tuning
has_fulltext	False
is_retracted	False
updated_date	2025-11-06T03:46:38.306776
primary_topic.id	https://openalex.org/T12205
primary_topic.field.id	https://openalex.org/fields/17
primary_topic.field.display_name	Computer Science
primary_topic.score	0.9332000017166138
primary_topic.domain.id	https://openalex.org/domains/3
primary_topic.domain.display_name	Physical Sciences
primary_topic.subfield.id	https://openalex.org/subfields/1711
primary_topic.subfield.display_name	Signal Processing
primary_topic.display_name	Time Series Analysis and Forecasting
related_works	https://openalex.org/W4391375266, https://openalex.org/W2899084033, https://openalex.org/W2748952813, https://openalex.org/W4390421286, https://openalex.org/W4280563792, https://openalex.org/W2140186469, https://openalex.org/W4389724018, https://openalex.org/W4318719684, https://openalex.org/W2775233965, https://openalex.org/W4318559728
cited_by_count	1
counts_by_year[0].year	2025
counts_by_year[0].cited_by_count	1
locations_count	1
best_oa_location.id	doi:10.1609/aaai.v39i15.33732
best_oa_location.is_oa	True
best_oa_location.source.id	https://openalex.org/S4210191458
best_oa_location.source.issn	2159-5399, 2374-3468
best_oa_location.source.type	conference
best_oa_location.source.is_oa	True
best_oa_location.source.issn_l	2159-5399
best_oa_location.source.is_core	False
best_oa_location.source.is_in_doaj	False
best_oa_location.source.display_name	Proceedings of the AAAI Conference on Artificial Intelligence
best_oa_location.source.host_organization	https://openalex.org/P4310320058
best_oa_location.source.host_organization_name	Association for the Advancement of Artificial Intelligence
best_oa_location.source.host_organization_lineage	https://openalex.org/P4310320058
best_oa_location.source.host_organization_lineage_names	Association for the Advancement of Artificial Intelligence
best_oa_location.license
best_oa_location.pdf_url	https://ojs.aaai.org/index.php/AAAI/article/download/33732/35887
best_oa_location.version	publishedVersion
best_oa_location.raw_type	journal-article
best_oa_location.license_id
best_oa_location.is_accepted	True
best_oa_location.is_published	True
best_oa_location.raw_source_name	Proceedings of the AAAI Conference on Artificial Intelligence
best_oa_location.landing_page_url	https://doi.org/10.1609/aaai.v39i15.33732
primary_location.id	doi:10.1609/aaai.v39i15.33732
primary_location.is_oa	True
primary_location.source.id	https://openalex.org/S4210191458
primary_location.source.issn	2159-5399, 2374-3468
primary_location.source.type	conference
primary_location.source.is_oa	True
primary_location.source.issn_l	2159-5399
primary_location.source.is_core	False
primary_location.source.is_in_doaj	False
primary_location.source.display_name	Proceedings of the AAAI Conference on Artificial Intelligence
primary_location.source.host_organization	https://openalex.org/P4310320058
primary_location.source.host_organization_name	Association for the Advancement of Artificial Intelligence
primary_location.source.host_organization_lineage	https://openalex.org/P4310320058
primary_location.source.host_organization_lineage_names	Association for the Advancement of Artificial Intelligence
primary_location.license
primary_location.pdf_url	https://ojs.aaai.org/index.php/AAAI/article/download/33732/35887
primary_location.version	publishedVersion
primary_location.raw_type	journal-article
primary_location.license_id
primary_location.is_accepted	True
primary_location.is_published	True
primary_location.raw_source_name	Proceedings of the AAAI Conference on Artificial Intelligence
primary_location.landing_page_url	https://doi.org/10.1609/aaai.v39i15.33732
publication_date	2025-04-11
publication_year	2025
referenced_works_count	0
abstract_inverted_index.a	4, 104, 161
abstract_inverted_index.FL	37, 73, 172, 177
abstract_inverted_index.HP	133, 163
abstract_inverted_index.In	99
abstract_inverted_index.ML	16, 26
abstract_inverted_index.an	149
abstract_inverted_index.at	135
abstract_inverted_index.in	10, 36, 61, 87, 198
abstract_inverted_index.is	3, 64, 70
abstract_inverted_index.of	160, 186
abstract_inverted_index.on	48, 169
abstract_inverted_index.to	24, 40, 92, 114, 126
abstract_inverted_index.we	102
abstract_inverted_index.FL,	88
abstract_inverted_index.FL.	62, 199
abstract_inverted_index.HPs	50, 94
abstract_inverted_index.ML,	53
abstract_inverted_index.Our	166
abstract_inverted_index.and	32, 139, 156, 174
abstract_inverted_index.are	38, 90
abstract_inverted_index.for	51, 72, 85, 95
abstract_inverted_index.the	28, 41, 93, 128, 137, 158, 170, 184, 187, 193
abstract_inverted_index.yet	118
abstract_inverted_index.(FL)	2
abstract_inverted_index.(HP)	43
abstract_inverted_index.(ML)	8
abstract_inverted_index.HPs,	129
abstract_inverted_index.This	63
abstract_inverted_index.been	83
abstract_inverted_index.both	136
abstract_inverted_index.have	82
abstract_inverted_index.some	80
abstract_inverted_index.they	89
abstract_inverted_index.this	100, 116
abstract_inverted_index.when	59
abstract_inverted_index.with	74, 143
abstract_inverted_index.While	79
abstract_inverted_index.data.	22
abstract_inverted_index.local	21, 30, 97
abstract_inverted_index.novel	105
abstract_inverted_index.prior	144
abstract_inverted_index.their	20, 67
abstract_inverted_index.these	54
abstract_inverted_index.train	15
abstract_inverted_index.types	134
abstract_inverted_index.vital	117
abstract_inverted_index.which	11, 130, 190
abstract_inverted_index.work,	101
abstract_inverted_index.yield	56
abstract_inverted_index.FedPop	121, 147
abstract_inverted_index.Tuning	112
abstract_inverted_index.called	108
abstract_inverted_index.client	29, 76, 96, 138
abstract_inverted_index.common	171
abstract_inverted_index.mainly	65
abstract_inverted_index.models	17
abstract_inverted_index.online	150
abstract_inverted_index.power.	78
abstract_inverted_index.search	164
abstract_inverted_index.server	33, 140
abstract_inverted_index.sides.	141
abstract_inverted_index.space.	165
abstract_inverted_index.tuning	49, 145
abstract_inverted_index.Despite	45
abstract_inverted_index.Non-IID	181
abstract_inverted_index.Similar	23
abstract_inverted_index.address	115
abstract_inverted_index.because	66
abstract_inverted_index.broader	162
abstract_inverted_index.clients	13
abstract_inverted_index.complex	175
abstract_inverted_index.employs	122, 148
abstract_inverted_index.limited	75, 91
abstract_inverted_index.machine	6
abstract_inverted_index.method,	189
abstract_inverted_index.methods	55, 197
abstract_inverted_index.propose	103
abstract_inverted_index.results	58
abstract_inverted_index.various	132
abstract_inverted_index.without	18
abstract_inverted_index.Compared	142
abstract_inverted_index.Learning	1
abstract_inverted_index.employed	60
abstract_inverted_index.enabling	157
abstract_inverted_index.learning	7
abstract_inverted_index.methods,	146
abstract_inverted_index.multiple	12
abstract_inverted_index.offering	153
abstract_inverted_index.optimize	127
abstract_inverted_index.problem.	120
abstract_inverted_index.proposed	84, 188
abstract_inverted_index.research	47
abstract_inverted_index.updates.	98
abstract_inverted_index.(FedPop),	113
abstract_inverted_index.Federated	0, 109
abstract_inverted_index.HP-Tuning	86
abstract_inverted_index.HP-tuning	106, 196
abstract_inverted_index.datasets,	178
abstract_inverted_index.empirical	167
abstract_inverted_index.extensive	46
abstract_inverted_index.framework	69
abstract_inverted_index.including	179
abstract_inverted_index.paradigm,	9
abstract_inverted_index.procedure	35
abstract_inverted_index.sensitive	39
abstract_inverted_index.algorithm,	107
abstract_inverted_index.algorithms	125
abstract_inverted_index.approaches	81
abstract_inverted_index.benchmarks	173
abstract_inverted_index.concurrent	194
abstract_inverted_index.efficiency	155
abstract_inverted_index.framework,	152
abstract_inverted_index.full-sized	180
abstract_inverted_index.pipelines,	27
abstract_inverted_index.real-world	176
abstract_inverted_index.selection.	44
abstract_inverted_index.suboptimal	57
abstract_inverted_index.unsuitable	71
abstract_inverted_index.validation	168
abstract_inverted_index.aggregation	34
abstract_inverted_index.centralized	52
abstract_inverted_index.challenging	119
abstract_inverted_index.computation	77
abstract_inverted_index.distributed	5
abstract_inverted_index.exploration	159
abstract_inverted_index.outperforms	192
abstract_inverted_index.ImageNet-1K,	182
abstract_inverted_index.accommodates	131
abstract_inverted_index.centralizing	19
abstract_inverted_index.conventional	25
abstract_inverted_index.demonstrates	183
abstract_inverted_index.evolutionary	124
abstract_inverted_index.optimization	31
abstract_inverted_index.computational	154
abstract_inverted_index.effectiveness	185
abstract_inverted_index.substantially	191
abstract_inverted_index.Hyperparameter	111
abstract_inverted_index.hyperparameter	42
abstract_inverted_index.collaboratively	14
abstract_inverted_index.Population-based	110
abstract_inverted_index.population-based	123
abstract_inverted_index.state-of-the-art	195
abstract_inverted_index."training-after-tuning"	68
abstract_inverted_index."tuning-while-training"	151
cited_by_percentile_year.max	95
cited_by_percentile_year.min	91
countries_distinct_count	0
institutions_distinct_count	4
citation_normalized_percentile.value	0.6433121
citation_normalized_percentile.is_in_top_1_percent	False
citation_normalized_percentile.is_in_top_10_percent	False