On the Fragility of Contribution Score Computation in Federated Learning
Not on the map. Why?
Abstract (arXiv)
This paper investigates the fragility of contribution evaluation in federated learning, a critical mechanism for ensuring fairness and incentivizing participation. We argue that contribution scores are susceptible to significant distortions from two fundamental perspectives: architectural sensitivity and intentional manipulation. First, we explore how different model aggregation methods impact these scores. While most research assumes a basic averaging approach, we demonstrate that advanced techniques, including those designed to handle unreliable or diverse clients, can unintentionally yet significantly alter the final scores. Second, we explore vulnerabilities posed by poisoning attacks, where malicious participants strategically manipulate their model updates to inflate their own contribution scores or reduce the importance of other participants. Through extensive experiments across diverse datasets and model architectures, implemented within the Flower framework, we rigorously show that both the choice of aggregation method and the presence of attackers are potent vectors for distorting contribution scores, highlighting a critical need for more robust evaluation schemes.
Key facts. Cited by 1 paper in the map (data through 7 Oct 2026). Builds on 43 of its 59 references that are in the map. First posted on arXiv 24 Sep 2025.
Builds on
43 of 59 references.
References · 43
- 2016Communication-Efficient Learning of Deep Networks from Decentralized Datacited by 7,592
- 2018Federated Optimization in Heterogeneous Networkscited by 2,971
- 2020Tackling the Objective Inconsistency Problem in Heterogeneous Federated Optimizationcited by 897
- 2018How To Backdoor Federated Learningcited by 745
- 2018Certified Robustness to Adversarial Examples with Differential Privacycited by 462
- 2020Flower: A Friendly Federated Learning Research Frameworkcited by 449
- 2018The Hidden Vulnerability of Distributed Learning in Byzantiumcited by 324
- 2020Data Poisoning Attacks Against Federated Learning Systemscited by 266
- 2018Machine Learning with Membership Privacy using Adversarial Regularizationcited by 177
- 2020Dataset Security for Machine Learning: Data Poisoning, Backdoor Attacks, and Defensescited by 171
- 2020Privacy and Robustness in Federated Learning: Attacks and Defensescited by 163
- 2018Mitigating Sybils in Federated Learning Poisoningcited by 158
31 more
- 2020To be Robust or to be Fair: Towards Fairness in Adversarial Trainingcited by 137
- 2019Privacy Risks of Securing Machine Learning Models against Adversarial Examplescited by 136
- 2022Data Banzhaf: A Robust Data Valuation Framework for Machine Learningcited by 126
- 2022The Shapley Value in Machine Learningcited by 97
- 2019Measure Contribution of Participants in Federated Learningcited by 78
- 2019Data Poisoning against Differentially-Private Learners: Attacks and Defensescited by 75
- 2021GTG-Shapley: Efficient and Accurate Participant Contribution Evaluation in Federated Learningcited by 73
- 2020Fast-Convergent Federated Learning with Adaptive Weightingcited by 67
- 2021Towards Fairness-Aware Federated Learningcited by 57
- 2020Fairness Through Robustness: Investigating Robustness Disparity in Deep Learningcited by 55
- 2023OpenDataVal: a Unified Benchmark for Data Valuationcited by 55
- 2020Evaluations and Methods for Explanation through Robustness Analysiscited by 51
- 2021Challenges and Approaches for Mitigating Byzantine Attacks in Federated Learningcited by 46
- 2020Robustness May Be at Odds with Fairness: An Empirical Study on Class-wise Accuracycited by 45
- 2021Evaluating Robustness of Counterfactual Explanationscited by 41
- 2023A Survey of Trustworthy Federated Learning with Perspectives on Security, Robustness, and Privacycited by 24
- 2022Industry-Scale Orchestrated Federated Learning for Drug Discoverycited by 19
- 2023Threshold KNN-Shapley: A Linear-Time and Privacy-Friendly Approach to Data Valuationcited by 19
- 2020Quality Inference in Federated Learning with Secure Aggregationcited by 15
- 2020An Exploratory Analysis on Users' Contributions in Federated Learningcited by 15
- 2022Secure Shapley Value for Cross-Silo Federated Learning (Technical Report)cited by 13
- 2023Stabilizing Estimates of Shapley Values with Control Variatescited by 9
- 2024Feature Inference Attack on Shapley Valuescited by 9
- 2022Differentially Private Shapley Values for Data Evaluationcited by 8
- 2021Is Shapley Value fair? Improving Client Selection for Mavericks in Federated Learningcited by 7
- 2023Towards Algorithmic Fairness by means of Instance-level Data Re-weighting based on Shapley Valuescited by 7
- 2022Reward Systems for Trustworthy Medical Federated Learningcited by 6
- 2025FedSV: Byzantine-Robust Federated Learning via Shapley Valuecited by 6
- 2024A Survey on Contribution Evaluation in Vertical Federated Learningcited by 5
- 2024ACE: A Model Poisoning Attack on Contribution Evaluation Methods in Federated Learningcited by 5
- 2024Privacy Implications of Explainable AI in Data-Driven Systemscited by 3
Built on by
Citing papers per year: 2026: 1.
1 paper.