L3S Best Publication of the Quarter (1/2024)
Category: Fairness
FairTrade: Achieving Pareto-Optimal Trade-offs Between Balanced Accuracy and Fairness in Federated Learning
Authors: Maryam Badar, Sandipan Sikdar, Wolfgang Nejdl and Marco Fisichella
Presented at the 38th Annual AAAI Conference on Artificial Intelligence (A* Conference)
https://ojs.aaai.org/index.php/AAAI/article/view/28971
Which problem does the research solve?
Most state-of-the-art fair Federated Learning methods report accuracy as the measure of performance, which can lead to misguided interpretations of the model’s effectiveness in mitigating discrimination. Our work attempts towards achieving Pareto-optimal trade-offs between balanced accuracy and fairness in a federated environment (FairTrade).
What is new about the research?
Our work is the first attempt towards the application of multi-objective optimization to negotiate the intricate balance between the model’s balanced accuracy and fairness.
Badar, M., Sikdar, S., Nejdl, W., & Fisichella, M. (2024). FairTrade: Achieving Pareto-Optimal Trade-Offs between Balanced Accuracy and Fairness in Federated Learning. Proceedings of the AAAI Conference on Artificial Intelligence, 38(10), 10962-10970. https://doi.org/10.1609/aaai.v38i10.28971
