# Fine-Tuning Personalization in Federated Learning to Mitigate Adversarial Clients Source: [Fine-Tuning Personalization in Federated Learning to Mitigate Adversarial Clients]() Collection index: [llms.txt]() Data updated: 2025-10-27T06:41:00.437Z Authors: Youssef Allouah, Abdellah El Mrini, Rachid Guerraoui, Nirupam Gupta, Rafael Pinot Published: 2024-09-30T14:31:19.000Z ## Abstract Federated learning (FL) is an appealing paradigm that allows a group of machines (a.k.a. clients) to learn collectively while keeping their data local. However, due to the heterogeneity between the clients' data distributions, the model obtained through the use of FL algorithms may perform poorly on some client's data. Personalization addresses this issue by enabling each client to have a different model tailored to their own data while simultaneously benefiting from the other clients' data. We consider an FL setting where some clients can be adversarial, and we derive conditions under which full collaboration fails. Specifically, we analyze the generalization performance of an interpolated personalized FL framework in the presence of adversarial clients, and we precisely characterize situations when full collaboration performs strictly worse than fine-tuned personalization. Our analysis determines how much we should scale down the level of collaboration, according to data heterogeneity and the tolerable fraction of adversarial clients. We support our findings with empirical results on mean estimation and binary classification problems, considering synthetic and benchmark image classification datasets. ## Recorded benchmark results A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. - [Image Classification on STL-10 (test) (Adversarial Robustness)](): Image Classification; STL-10 (test); Accuracy (Benign): 37.95. - [Image Classification on CIFAR-10 (Adversarial Robustness FGSM, PGD, SparseFool, PAP)](): Image Classification; CIFAR-10; Benign Accuracy: 39.96. - [Image Classification on CIFAR-100 (Adversarial Robustness: BA, FGSM, PGD, SparseFool, PAP)](): Image Classification; CIFAR-100; Benign Accuracy (BA): 14.65. - [Image Classification Robustness on STL-10 (Adversarial Attacks)](): Image Classification; STL-10; Accuracy (Benign): 32.92. - [Image Classification on GTSRB (test)](): Image Classification; GTSRB (test); Accuracy (Clean): 65.52. - [Robust Image Classification on GTSRB](): Image Classification; GTSRB; Natural Accuracy: 60.73. ## Original sources and implementation [Original paper]() [PDF]()