Robust Server Defense Against Unreliable Clients in One-Shot Fair Collaborative Machine Learning
For practitioners of collaborative machine learning, this work provides a robust defense against fairness degradation caused by unreliable clients in one-shot settings.
This paper addresses the vulnerability of one-shot collaborative machine learning to unreliable clients who provide biased proxy data, which can degrade fairness. The proposed server-side defense framework using bilevel optimization and a small trusted root dataset improves fairness with little accuracy loss, even when unreliable clients are the majority.
Collaborative machine learning (CML) enables multiple clients to train a global model jointly in a data-distributed setting. To address data privacy and communication efficiency, one-shot CML has been increasingly adopted, where clients communicate with the server only once by sharing synthetic or processed proxy data. This single-round communication, however, eliminates the possibility of iterative correction at the server, making the learning process particularly vulnerable to client unreliability. In this setting, unreliable clients, whether malicious or non-malicious, may provide biased proxy data that favors certain groups, thereby degrading the fairness of the global model and harming minority or unprivileged groups. In this work, we propose a server-side defense framework based on a bilevel optimization formulation. The proposed approach learns client-level weights to mitigate the influence of biased client proxy data while enforcing fairness constraints by using a very small trusted root dataset available at the server. Experimental results on benchmark datasets show that our method improves fairness with little accuracy loss under biased proxy data contributions from unreliable clients. Moreover, the proposed approach remains effective even when unreliable clients make up a majority of the system, consistently outperforming other existing methods.