Details
| Original language | English |
|---|---|
| Pages (from-to) | 2761-2786 |
| Number of pages | 26 |
| Journal | Machine learning |
| Volume | 112 |
| Issue number | 8 |
| Early online date | 26 Jul 2023 |
| Publication status | Published - Aug 2023 |
Abstract
Federated learning is an emerging collaborative learning paradigm of Machine learning involving distributed and heterogeneous clients. Enormous collections of continuously arriving heterogeneous data residing on distributed clients require federated adaptation of efficient mining algorithms to enable fair and high-quality predictions with privacy guarantees and minimal response delay. In this context, we propose a federated adaptation that mitigates discrimination embedded in the streaming data while handling concept drifts (FAC-Fed). We present a novel adaptive data augmentation method that mitigates client-side discrimination embedded in the data during optimization, resulting in an optimized and fair centralized server. Extensive experiments on a set of publicly available streaming and static datasets confirm the effectiveness of the proposed method. To the best of our knowledge, this work is the first attempt towards fairness-aware federated adaptation for stream classification, therefore, to prove the superiority of our proposed method over state-of-the-art, we compare the centralized version of our proposed method with three centralized stream classification baseline models (FABBOO, FAHT, CSMOTE). The experimental results show that our method outperforms the current methods in terms of both discrimination mitigation and predictive performance.
Keywords
- Deep neural network, Fairness, Federated learning, Hedge backpropagation, Privacy, Stream classification
ASJC Scopus subject areas
- Computer Science(all)
- Software
- Computer Science(all)
- Artificial Intelligence
Cite this
- Standard
- Harvard
- Apa
- Vancouver
- BibTeX
- RIS
In: Machine learning, Vol. 112, No. 8, 08.2023, p. 2761-2786.
Research output: Contribution to journal › Article › Research › peer review
}
TY - JOUR
T1 - FAC-fed
T2 - Federated adaptation for fairness and concept drift aware stream classification
AU - Badar, Maryam
AU - Nejdl, Wolfgang
AU - Fisichella, Marco
N1 - Funding Information: Open Access funding enabled and organized by Projekt DEAL. The Lower Saxony Ministry of Science and Culture (Niedersächsische Ministerium für Wissenschaft und Kultur). This research was partially funded by the Lower Saxony Ministry of Science and Culture, Germany (Niedersächsische Ministerium für Wissenschaft und Kultur) and the Federal Ministry of Education and Research (BMBF), Germany under the project LeibnizKILabor with grant No. 01DD20003.
PY - 2023/8
Y1 - 2023/8
N2 - Federated learning is an emerging collaborative learning paradigm of Machine learning involving distributed and heterogeneous clients. Enormous collections of continuously arriving heterogeneous data residing on distributed clients require federated adaptation of efficient mining algorithms to enable fair and high-quality predictions with privacy guarantees and minimal response delay. In this context, we propose a federated adaptation that mitigates discrimination embedded in the streaming data while handling concept drifts (FAC-Fed). We present a novel adaptive data augmentation method that mitigates client-side discrimination embedded in the data during optimization, resulting in an optimized and fair centralized server. Extensive experiments on a set of publicly available streaming and static datasets confirm the effectiveness of the proposed method. To the best of our knowledge, this work is the first attempt towards fairness-aware federated adaptation for stream classification, therefore, to prove the superiority of our proposed method over state-of-the-art, we compare the centralized version of our proposed method with three centralized stream classification baseline models (FABBOO, FAHT, CSMOTE). The experimental results show that our method outperforms the current methods in terms of both discrimination mitigation and predictive performance.
AB - Federated learning is an emerging collaborative learning paradigm of Machine learning involving distributed and heterogeneous clients. Enormous collections of continuously arriving heterogeneous data residing on distributed clients require federated adaptation of efficient mining algorithms to enable fair and high-quality predictions with privacy guarantees and minimal response delay. In this context, we propose a federated adaptation that mitigates discrimination embedded in the streaming data while handling concept drifts (FAC-Fed). We present a novel adaptive data augmentation method that mitigates client-side discrimination embedded in the data during optimization, resulting in an optimized and fair centralized server. Extensive experiments on a set of publicly available streaming and static datasets confirm the effectiveness of the proposed method. To the best of our knowledge, this work is the first attempt towards fairness-aware federated adaptation for stream classification, therefore, to prove the superiority of our proposed method over state-of-the-art, we compare the centralized version of our proposed method with three centralized stream classification baseline models (FABBOO, FAHT, CSMOTE). The experimental results show that our method outperforms the current methods in terms of both discrimination mitigation and predictive performance.
KW - Deep neural network
KW - Fairness
KW - Federated learning
KW - Hedge backpropagation
KW - Privacy
KW - Stream classification
UR - http://www.scopus.com/inward/record.url?scp=85165872586&partnerID=8YFLogxK
U2 - 10.1007/s10994-023-06360-7
DO - 10.1007/s10994-023-06360-7
M3 - Article
AN - SCOPUS:85165872586
VL - 112
SP - 2761
EP - 2786
JO - Machine learning
JF - Machine learning
SN - 0885-6125
IS - 8
ER -