Details
| Original language | English |
|---|---|
| Title of host publication | Machine Learning and Knowledge Discovery in Databases |
| Subtitle of host publication | European Conference, ECML PKDD 2022, Grenoble, France, September 19–23, 2022, Proceedings, Part I |
| Editors | Massih-Reza Amini, Stéphane Canu, Asja Fischer, Tias Guns, Petra Kralj Novak, Grigorios Tsoumakas |
| Place of Publication | Cham |
| Pages | 710-726 |
| Number of pages | 17 |
| Edition | 1. |
| ISBN (electronic) | 978-3-031-26387-3 |
| Publication status | Published - 17 Mar 2023 |
| Event | 22nd Joint European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases, ECML PKDD 2022 - Grenoble, France Duration: 19 Sept 2022 → 23 Sept 2022 |
Publication series
| Name | Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) |
|---|---|
| Volume | 13713 LNAI |
| ISSN (Print) | 0302-9743 |
| ISSN (electronic) | 1611-3349 |
Abstract
Fairness-aware learning mainly focuses on single task learning (STL). The fairness implications of multi-task learning (MTL) have only recently been considered and a seminal approach has been proposed that considers the fairness-accuracy trade-off for each task and the performance trade-off among different tasks. Instead of a rigid fairness-accuracy trade-off formulation, we propose a flexible approach that learns how to be fair in a MTL setting by selecting which objective (accuracy or fairness) to optimize at each step. We introduce the L2T-FMT algorithm that is a teacher-student network trained collaboratively; the student learns to solve the fair MTL problem while the teacher instructs the student to learn from either accuracy or fairness, depending on what is harder to learn for each task. Moreover, this dynamic selection of which objective to use at each step for each task reduces the number of trade-off weights from 2T to T, where T is the number of tasks. Our experiments on three real datasets show that L2T-FMT improves on both fairness (12–19%) and accuracy (up to 2%) over state-of-the-art approaches.
ASJC Scopus subject areas
- Mathematics(all)
- Theoretical Computer Science
- Computer Science(all)
Cite this
- Standard
- Harvard
- Apa
- Vancouver
- BibTeX
- RIS
Machine Learning and Knowledge Discovery in Databases : European Conference, ECML PKDD 2022, Grenoble, France, September 19–23, 2022, Proceedings, Part I. ed. / Massih-Reza Amini; Stéphane Canu; Asja Fischer; Tias Guns; Petra Kralj Novak; Grigorios Tsoumakas. 1. ed. Cham, 2023. p. 710-726 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 13713 LNAI).
Research output: Chapter in book/report/conference proceeding › Conference contribution › Research › peer review
}
TY - GEN
T1 - Learning to Teach Fairness-Aware Deep Multi-task Learning
AU - Roy, Arjun
AU - Ntoutsi, Eirini
N1 - Funding Information: Acknowledgements. The work of A.Roy is supported by Volkswagen Foundation under the call “Artificial Intelligence and the Society of the Future” project “Bias and Discrimination in Big Data and Algorithmic Processing - BIAS”.
PY - 2023/3/17
Y1 - 2023/3/17
N2 - Fairness-aware learning mainly focuses on single task learning (STL). The fairness implications of multi-task learning (MTL) have only recently been considered and a seminal approach has been proposed that considers the fairness-accuracy trade-off for each task and the performance trade-off among different tasks. Instead of a rigid fairness-accuracy trade-off formulation, we propose a flexible approach that learns how to be fair in a MTL setting by selecting which objective (accuracy or fairness) to optimize at each step. We introduce the L2T-FMT algorithm that is a teacher-student network trained collaboratively; the student learns to solve the fair MTL problem while the teacher instructs the student to learn from either accuracy or fairness, depending on what is harder to learn for each task. Moreover, this dynamic selection of which objective to use at each step for each task reduces the number of trade-off weights from 2T to T, where T is the number of tasks. Our experiments on three real datasets show that L2T-FMT improves on both fairness (12–19%) and accuracy (up to 2%) over state-of-the-art approaches.
AB - Fairness-aware learning mainly focuses on single task learning (STL). The fairness implications of multi-task learning (MTL) have only recently been considered and a seminal approach has been proposed that considers the fairness-accuracy trade-off for each task and the performance trade-off among different tasks. Instead of a rigid fairness-accuracy trade-off formulation, we propose a flexible approach that learns how to be fair in a MTL setting by selecting which objective (accuracy or fairness) to optimize at each step. We introduce the L2T-FMT algorithm that is a teacher-student network trained collaboratively; the student learns to solve the fair MTL problem while the teacher instructs the student to learn from either accuracy or fairness, depending on what is harder to learn for each task. Moreover, this dynamic selection of which objective to use at each step for each task reduces the number of trade-off weights from 2T to T, where T is the number of tasks. Our experiments on three real datasets show that L2T-FMT improves on both fairness (12–19%) and accuracy (up to 2%) over state-of-the-art approaches.
UR - http://www.scopus.com/inward/record.url?scp=85151065743&partnerID=8YFLogxK
U2 - 10.48550/arXiv.2206.08403
DO - 10.48550/arXiv.2206.08403
M3 - Conference contribution
AN - SCOPUS:85151065743
SN - 9783031263866
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 710
EP - 726
BT - Machine Learning and Knowledge Discovery in Databases
A2 - Amini, Massih-Reza
A2 - Canu, Stéphane
A2 - Fischer, Asja
A2 - Guns, Tias
A2 - Kralj Novak, Petra
A2 - Tsoumakas, Grigorios
CY - Cham
T2 - 22nd Joint European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases, ECML PKDD 2022
Y2 - 19 September 2022 through 23 September 2022
ER -