Unbalanced Data Supported by Federated Learning with Uncertainty by Different Aggregation Methods

Barbara Pekala; Anna Wilbik; Krzysztof Dyczkowski; Jaroslaw Szkola

doi:10.1007/978-3-031-99178-3_7

Unbalanced Data Supported by Federated Learning with Uncertainty by Different Aggregation Methods

Barbara Pekala^*
, Anna Wilbik
, Krzysztof Dyczkowski
, Jaroslaw Szkola

^*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceeding › Conference article in proceeding › Academic › peer-review

Abstract

Federated learning enables multiple clients to collaboratively train a machine learning model while keeping their local data private-omitting sharing data, making it a privacy-preserving technique. Data plays a key role in these models. However, in some cases, a single organization may not have enough data or high-quality data to build a reliable model, especially in a rapidly changing environment. In horizontal federated learning, each organization/client continuously refines its model, which is periodically fused and distributed among all participating clients in the federation for further enhancement. The fusion/aggregation process typically relies on a weighted averaging approach, where the weights are determined by the quality of each client’s model. This study explores approaches by using federated learning with respect to uncertainty and examines various aggregation strategies based on the performance of local models.

Original language	English
Title of host publication	Uncertainty and Imprecision in Decision Making and Decision Support - New Advances, Challenges, and Perspectives - Selected Papers from the IWIFSGN-2023 - 21st International Workshop on Intuitionistic Fuzzy Sets and Generalized Nets, 2023
Editors	Krassimir T. Atanassov, Vassia Atanassova, Janusz Kacprzyk, Andrzej Kaluszko, Jan Owsinski, Eulalia Szmidt, Maciej Krawczak, Sotir N. Sotirov, Evdokia Sotirova, Slawomir Zadrozny
Publisher	Springer
Pages	65-83
Number of pages	19
Volume	1550 LNNS
ISBN (Print)	9783031991776
DOIs	https://doi.org/10.1007/978-3-031-99178-3_7
Publication status	Published - 2026
Event	21st International Workshop on Intuitionistic Fuzzy Sets and Generalized Nets - Warsaw, Poland Duration: 20 Oct 2023 → 20 Oct 2023 Conference number: 21 https://www2.ibspan.waw.pl/ifs2023/

Publication series

Series	Lecture Notes in Networks and Systems
Volume	1550 LNNS
ISSN	2367-3370

Conference

Conference	21st International Workshop on Intuitionistic Fuzzy Sets and Generalized Nets
Abbreviated title	IWIFSGN 2023
Country/Territory	Poland
City	Warsaw
Period	20/10/23 → 20/10/23
Internet address	https://www2.ibspan.waw.pl/ifs2023/

Keywords

Aggregation process
Federated learning
Uncertainty data

Access to Document

10.1007/978-3-031-99178-3_7

Cite this

Pekala, B., Wilbik, A., Dyczkowski, K., & Szkola, J. (2026). Unbalanced Data Supported by Federated Learning with Uncertainty by Different Aggregation Methods. In K. T. Atanassov, V. Atanassova, J. Kacprzyk, A. Kaluszko, J. Owsinski, E. Szmidt, M. Krawczak, S. N. Sotirov, E. Sotirova, & S. Zadrozny (Eds.), Uncertainty and Imprecision in Decision Making and Decision Support - New Advances, Challenges, and Perspectives - Selected Papers from the IWIFSGN-2023 - 21st International Workshop on Intuitionistic Fuzzy Sets and Generalized Nets, 2023 (Vol. 1550 LNNS, pp. 65-83). Springer. https://doi.org/10.1007/978-3-031-99178-3_7

Pekala, Barbara ; Wilbik, Anna ; Dyczkowski, Krzysztof et al. / Unbalanced Data Supported by Federated Learning with Uncertainty by Different Aggregation Methods. Uncertainty and Imprecision in Decision Making and Decision Support - New Advances, Challenges, and Perspectives - Selected Papers from the IWIFSGN-2023 - 21st International Workshop on Intuitionistic Fuzzy Sets and Generalized Nets, 2023. editor / Krassimir T. Atanassov ; Vassia Atanassova ; Janusz Kacprzyk ; Andrzej Kaluszko ; Jan Owsinski ; Eulalia Szmidt ; Maciej Krawczak ; Sotir N. Sotirov ; Evdokia Sotirova ; Slawomir Zadrozny. Vol. 1550 LNNS Springer, 2026. pp. 65-83 (Lecture Notes in Networks and Systems, Vol. 1550 LNNS).

@inproceedings{89cd0699927f459ba77ba2fe6457ff6f,

title = "Unbalanced Data Supported by Federated Learning with Uncertainty by Different Aggregation Methods",

abstract = "Federated learning enables multiple clients to collaboratively train a machine learning model while keeping their local data private-omitting sharing data, making it a privacy-preserving technique. Data plays a key role in these models. However, in some cases, a single organization may not have enough data or high-quality data to build a reliable model, especially in a rapidly changing environment. In horizontal federated learning, each organization/client continuously refines its model, which is periodically fused and distributed among all participating clients in the federation for further enhancement. The fusion/aggregation process typically relies on a weighted averaging approach, where the weights are determined by the quality of each client{\textquoteright}s model. This study explores approaches by using federated learning with respect to uncertainty and examines various aggregation strategies based on the performance of local models.",

keywords = "Aggregation process, Federated learning, Uncertainty data",

author = "Barbara Pekala and Anna Wilbik and Krzysztof Dyczkowski and Jaroslaw Szkola",

note = "Publisher Copyright: {\textcopyright} The Author(s), under exclusive license to Springer Nature Switzerland AG 2026.; 21st International Workshop on Intuitionistic Fuzzy Sets and Generalized Nets, IWIFSGN 2023 ; Conference date: 20-10-2023 Through 20-10-2023",

year = "2026",

doi = "10.1007/978-3-031-99178-3\_7",

language = "English",

isbn = "9783031991776",

volume = "1550 LNNS",

series = "Lecture Notes in Networks and Systems",

publisher = "Springer",

pages = "65--83",

editor = "Atanassov, \{Krassimir T.\} and Vassia Atanassova and Janusz Kacprzyk and Andrzej Kaluszko and Jan Owsinski and Eulalia Szmidt and Maciej Krawczak and Sotirov, \{Sotir N.\} and Evdokia Sotirova and Slawomir Zadrozny",

booktitle = "Uncertainty and Imprecision in Decision Making and Decision Support - New Advances, Challenges, and Perspectives - Selected Papers from the IWIFSGN-2023 - 21st International Workshop on Intuitionistic Fuzzy Sets and Generalized Nets, 2023",

address = "United States",

url = "https://www2.ibspan.waw.pl/ifs2023/",

}

Pekala, B, Wilbik, A, Dyczkowski, K & Szkola, J 2026, Unbalanced Data Supported by Federated Learning with Uncertainty by Different Aggregation Methods. in KT Atanassov, V Atanassova, J Kacprzyk, A Kaluszko, J Owsinski, E Szmidt, M Krawczak, SN Sotirov, E Sotirova & S Zadrozny (eds), Uncertainty and Imprecision in Decision Making and Decision Support - New Advances, Challenges, and Perspectives - Selected Papers from the IWIFSGN-2023 - 21st International Workshop on Intuitionistic Fuzzy Sets and Generalized Nets, 2023. vol. 1550 LNNS, Springer, Lecture Notes in Networks and Systems, vol. 1550 LNNS, pp. 65-83, 21st International Workshop on Intuitionistic Fuzzy Sets and Generalized Nets, Warsaw, Poland, 20/10/23. https://doi.org/10.1007/978-3-031-99178-3_7

Unbalanced Data Supported by Federated Learning with Uncertainty by Different Aggregation Methods. / Pekala, Barbara; Wilbik, Anna; Dyczkowski, Krzysztof et al.
Uncertainty and Imprecision in Decision Making and Decision Support - New Advances, Challenges, and Perspectives - Selected Papers from the IWIFSGN-2023 - 21st International Workshop on Intuitionistic Fuzzy Sets and Generalized Nets, 2023. ed. / Krassimir T. Atanassov; Vassia Atanassova; Janusz Kacprzyk; Andrzej Kaluszko; Jan Owsinski; Eulalia Szmidt; Maciej Krawczak; Sotir N. Sotirov; Evdokia Sotirova; Slawomir Zadrozny. Vol. 1550 LNNS Springer, 2026. p. 65-83 (Lecture Notes in Networks and Systems, Vol. 1550 LNNS).

Research output: Chapter in Book/Report/Conference proceeding › Conference article in proceeding › Academic › peer-review

TY - GEN

T1 - Unbalanced Data Supported by Federated Learning with Uncertainty by Different Aggregation Methods

AU - Pekala, Barbara

AU - Wilbik, Anna

AU - Dyczkowski, Krzysztof

AU - Szkola, Jaroslaw

N1 - Conference code: 21

PY - 2026

Y1 - 2026

N2 - Federated learning enables multiple clients to collaboratively train a machine learning model while keeping their local data private-omitting sharing data, making it a privacy-preserving technique. Data plays a key role in these models. However, in some cases, a single organization may not have enough data or high-quality data to build a reliable model, especially in a rapidly changing environment. In horizontal federated learning, each organization/client continuously refines its model, which is periodically fused and distributed among all participating clients in the federation for further enhancement. The fusion/aggregation process typically relies on a weighted averaging approach, where the weights are determined by the quality of each client’s model. This study explores approaches by using federated learning with respect to uncertainty and examines various aggregation strategies based on the performance of local models.

AB - Federated learning enables multiple clients to collaboratively train a machine learning model while keeping their local data private-omitting sharing data, making it a privacy-preserving technique. Data plays a key role in these models. However, in some cases, a single organization may not have enough data or high-quality data to build a reliable model, especially in a rapidly changing environment. In horizontal federated learning, each organization/client continuously refines its model, which is periodically fused and distributed among all participating clients in the federation for further enhancement. The fusion/aggregation process typically relies on a weighted averaging approach, where the weights are determined by the quality of each client’s model. This study explores approaches by using federated learning with respect to uncertainty and examines various aggregation strategies based on the performance of local models.

KW - Aggregation process

KW - Federated learning

KW - Uncertainty data

U2 - 10.1007/978-3-031-99178-3_7

DO - 10.1007/978-3-031-99178-3_7

M3 - Conference article in proceeding

SN - 9783031991776

VL - 1550 LNNS

T3 - Lecture Notes in Networks and Systems

SP - 65

EP - 83

BT - Uncertainty and Imprecision in Decision Making and Decision Support - New Advances, Challenges, and Perspectives - Selected Papers from the IWIFSGN-2023 - 21st International Workshop on Intuitionistic Fuzzy Sets and Generalized Nets, 2023

A2 - Atanassov, Krassimir T.

A2 - Atanassova, Vassia

A2 - Kacprzyk, Janusz

A2 - Kaluszko, Andrzej

A2 - Owsinski, Jan

A2 - Szmidt, Eulalia

A2 - Krawczak, Maciej

A2 - Sotirov, Sotir N.

A2 - Sotirova, Evdokia

A2 - Zadrozny, Slawomir

PB - Springer

T2 - 21st International Workshop on Intuitionistic Fuzzy Sets and Generalized Nets

Y2 - 20 October 2023 through 20 October 2023

ER -

Pekala B, Wilbik A, Dyczkowski K, Szkola J. Unbalanced Data Supported by Federated Learning with Uncertainty by Different Aggregation Methods. In Atanassov KT, Atanassova V, Kacprzyk J, Kaluszko A, Owsinski J, Szmidt E, Krawczak M, Sotirov SN, Sotirova E, Zadrozny S, editors, Uncertainty and Imprecision in Decision Making and Decision Support - New Advances, Challenges, and Perspectives - Selected Papers from the IWIFSGN-2023 - 21st International Workshop on Intuitionistic Fuzzy Sets and Generalized Nets, 2023. Vol. 1550 LNNS. Springer. 2026. p. 65-83. (Lecture Notes in Networks and Systems, Vol. 1550 LNNS). doi: 10.1007/978-3-031-99178-3_7

Unbalanced Data Supported by Federated Learning with Uncertainty by Different Aggregation Methods

Abstract

Publication series

Conference

Keywords

Access to Document

Fingerprint

Cite this