Detect & Score: Privacy-Preserving Misbehaviour Detection and Contribution Evaluation in Federated Learning

2025/06/30 by Marvin Xhemrishi, Xhemrishi, Marvin, Alexandre Graell i Amat +3
Computer Science · Decision Sciences · #Advanced Graph Neural Networks #Cryptography and Security (cs.CR) #Data Quality and Management #Distributed #FOS: Computer and information sciences #Machine Learning (cs.LG) #Parallel #Privacy-Preserving Technologies in Data #and Cluster Computing (cs.DC)

paper · pdf · doi:10.48550/arxiv.2506.23583

openalex publication_date 2025/06/30 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

Federated learning with secure aggregation enables private and collaborative learning from decentralised data without leaking sensitive client information. However, secure aggregation also complicates the detection of malicious client behaviour and the evaluation of individual client contributions to the learning. To address these challenges, QI (Pejo et al.) and FedGT (Xhemrishi et al.) were proposed for contribution evaluation (CE) and misbehaviour detection (MD), respectively. QI, however, lacks adequate MD accuracy due to its reliance on the random selection of clients in each training round, while FedGT lacks the CE ability. In this work, we combine the strengths of QI and FedGT to achieve both robust MD and accurate CE. Our experiments demonstrate superior performance compared to using either method independently.

Citations

• Label Poisoning is All You Need
• OpenDataVal: a Unified Benchmark for Data Valuation
• FedGT: Identification of Malicious Clients in Federated Learning with Secure Aggregation
• Learning to Invert: Simple Adaptive Attacks for Gradient Inversion in Federated Learning
• FLamby: Datasets and Benchmarks for Cross-Silo Federated Learning in Realistic Healthcare Settings
• Autoregressive Perturbations for Data Poisoning
• Differentially Private Shapley Values for Data Evaluation
• Defense Strategies Toward Model Poisoning Attacks in Federated Learning: A Survey
• The Shapley Value in Machine Learning
• Fast-Convergent Federated Learning with Adaptive Weighting
• An Exploratory Analysis on Users' Contributions in Federated Learning
• A Principled Approach to Data Valuation for Federated Learning
• Byzantine-Resilient Secure Federated Learning
• Backdoor Learning: A Survey
• Data Poisoning Attacks Against Federated Learning Systems
• Quality Inference in Federated Learning with Secure Aggregation
• Robust Aggregation for Federated Learning
• Advances and Open Problems in Federated Learning
• Measure Contribution of Participants in Federated Learning
• Deep Leakage from Gradients
• Mitigating Sybils in Federated Learning Poisoning
• How To Backdoor Federated Learning
• Exploiting Unintended Feature Leakage in Collaborative Learning
• A Survey on Homomorphic Encryption Schemes: Theory and Implementation
• Stealing Machine Learning Models via Prediction APIs
• Beyond Accuracy: What Data Quality Means to Data Consumers

Related

• FedGT: Identification of Malicious Clients in Federated Learning with Secure Aggregation
• Secure Aggregation Is Not All You Need: Mitigating Privacy Attacks with Noise Tolerance in Federated Learning
• ACE: A Model Poisoning Attack on Contribution Evaluation Methods in Federated Learning
• FedBayes: A Zero-Trust Federated Learning Aggregation to Defend Against Adversarial Attacks
• An Exploratory Analysis on Users' Contributions in Federated Learning