mlsanityData to 14 Oct 2026

Byzantine-Robust Federated Learning: Impact of Client Subsampling and Local Updates

arXiv 2402.12780 ↗First posted 20 Feb 2024cs.LGcited by 14

Not on the map. Why?

Abstract (arXiv)

The possibility of adversarial (a.k.a., em Byzantine) clients makes federated learning (FL) prone to arbitrary manipulation. The natural approach to robustify FL against adversarial clients is to replace the simple averaging operation at the server in the standard mathsfFedAvg algorithm by a robust averaging rule. While a significant amount of work has been devoted to studying the convergence of federated em robust averaging (which we denote by mathsfFedRo), prior work has largely ignored the impact of em client subsampling and em local steps, two fundamental FL characteristics. While client subsampling increases the effective fraction of Byzantine clients, local steps increase the drift between the local updates computed by honest (i.e., non-Byzantine) clients. Consequently, a careless deployment of mathsfFedRo could yield poor performance. We validate this observation by presenting an in-depth analysis of mathsfFedRo tightly analyzing the impact of client subsampling and local steps. Specifically, we present a sufficient condition on client subsampling for nearly-optimal convergence of mathsfFedRo (for smooth non-convex loss). Also, we show that the rate of improvement in learning accuracy em diminishes with respect to the number of clients subsampled, as soon as the sample size exceeds a threshold value. Interestingly, we also observe that under a careful choice of step-sizes, the learning error due to Byzantine clients decreases with the number of local steps. We validate our theory by experiments on the FEMNIST and CIFAR-10 image classification tasks.

Key facts. Cited by 14 papers in the map (data through 14 Oct 2026). Builds on 25 of its 39 references that are in the map. First posted on arXiv 20 Feb 2024.

Builds on

25 of 39 references.

References · 25

  1. 2016Communication-Efficient Learning of Deep Networks from Decentralized Datacited by 7,669
  2. 2019Advances and Open Problems in Federated Learningcited by 3,046
  3. 2016Optimization Methods for Large-Scale Machine Learningcited by 1,150
  4. 2020Tackling the Objective Inconsistency Problem in Heterogeneous Federated Optimizationcited by 906
  5. 2018LEAF: A Benchmark for Federated Settingscited by 792
  6. 2013Stochastic First- and Zeroth-order Methods for Nonconvex Stochastic Programmingcited by 724
  7. 2018Byzantine-Robust Distributed Learning: Towards Optimal Statistical Ratescited by 711
  8. 2019Robust Aggregation for Federated Learningcited by 317
  9. 2019A Little Is Enough: Circumventing Defenses For Distributed Learningcited by 288
  10. 2021Achieving Linear Speedup with Partial Worker Participation in Non-IID Federated Learningcited by 193
  11. 2019Fall of Empires: Breaking Byzantine-tolerant SGD by Inner Product Manipulationcited by 141
  12. 2020Learning from History for Byzantine Robust Optimizationcited by 137
13 more
  1. 2020Byzantine-Robust Learning on Heterogeneous Datasets via Bucketingcited by 128
  2. 2018Generalized Byzantine-tolerant SGDcited by 95
  3. 2023Fixing by Mixing: A Recipe for Optimal Byzantine ML under Heterogeneitycited by 65
  4. 2022FedVARP: Tackling the Variance Due to Partial Client Participation in Federated Learningcited by 64
  5. 2022Byzantine Machine Learning Made Easy by Resilient Averaging of Momentumscited by 63
  6. 2020Byzantine-Resilient Non-Convex Stochastic Gradient Descentcited by 57
  7. 2022Byzantine-Robust Federated Learning with Optimal Statistical Rates and Privacy Guaranteescited by 30
  8. 2021Approximate Byzantine Fault-Tolerance in Distributed Optimizationcited by 28
  9. 2020Byzantine-Resilient High-Dimensional Federated Learningcited by 27
  10. 2022Personalization Improves Privacy-Accuracy Tradeoffs in Federated Learningcited by 23
  11. 2021Robust Training in High Dimensions via Block Coordinate Geometric Median Descentcited by 20
  12. 2021On the Strategyproofness of the Geometric Mediancited by 9
  13. 2021Byzantine Fault-Tolerance in Federated Local SGD under 2f-Redundancycited by 7

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 9; 2026: 5.

14 papers.

Sort:

  1. 2025ByzFL: Research Framework for Robust Federated Learningcited by 4
  2. 2025Bant: Byzantine Antidote via Trial Function and Trust Scorescited by 3
  3. 2025FedGreed: A Byzantine-Robust Loss-Based Aggregation Method for Federated Learningcited by 3
  4. 2025Delayed Momentum Aggregation: Communication-efficient Byzantine-robust Federated Learning with Partial Participationcited by 3
  5. 2025Robust Federated Learning under Adversarial Attacks via Loss-Based Client Clusteringcited by 2
  6. 2025Federated Generalised Variational Inference: A Robust Probabilistic Federated Learning Frameworkcited by 1
  7. 2026On the Gradient Heterogeneity Dynamics of Adversarially Robust Federated Regressioncited by 1
  8. 2025Analysis of Asynchronous Federated Learning: Unraveling the Interactions between Gradient Compression, Delay, and Data Heterogeneitycited by 0
  9. 2025Trade-off in Estimating the Number of Byzantine Clients in Federated Learningcited by 0
  10. 2025Robust and Efficient Collaborative Learningcited by 0
  11. 2026Revealing the influence of participant failures on model quality in cross-silo Federated Learningcited by 0
  12. 2026Theoretical Foundations of Communication-Efficient, Robust, and Practical Distributed and Federated Optimizationcited by 0
2 more
  1. 2026FedVAR: Prototype-Aligned Federated Framework for Video Anomaly Recognitioncited by 0
  2. 2026Robustifying Asynchronous SGD via Soft Throttlingcited by 0