paper-with-me

Papers

Reconciling Communication Compression and Byzantine-Robustness in Distributed Learning

2025-08-23 · Diksha Gupta, Antonio Honsell, Chuan Xu, Nirupam Gupta, Giovanni Neglia arxiv

Distributed learning enables scalable model training over decentralized data, but remains hindered by Byzantine faults and high communication costs. While both challenges have been studied extensively in isolation, their interplay has received limited attention. Prior work has shown that naively combining communication compression with Byzantine-robust aggregation can severely weaken resilience to faulty nodes. The current state-of-the-art, Byz-DASHA-PAGE, leverages a momentum-based variance reduction scheme to counteract the negative effect of compression noise on Byzantine robustness. In this work, we introduce RoSDHB, a new algorithm that integrates classical Polyak momentum with a coordinated compression strategy. Theoretically, RoSDHB matches the convergence guarantees of Byz-DASHA-PAGE under the standard $(G,B)$-gradient dissimilarity model, while relying on milder assumptions and requiring less memory and communication per client. Empirically, RoSDHB demonstrates stronger robustness while achieving substantial communication savings compared to Byz-DASHA-PAGE.

📄 PDF Abstract BibTeX arXiv:2508.17129

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

Communication Compression for Byzantine Robust Learning: New Efficient Algorithms and Improved Rates

2023-10-15 · Ahmad Rammal, Kaja Gruntkowska, Nikita Fedin, Eduard Gorbunov 외

Byzantine robustness is an essential feature of algorithms for certain distributed optimization problems, typically encountered in collaborative/federated learning. These problems are usually huge-scale, implying that co…

Distributed OptimizationFederated Learning

Byzantine-Robust and Communication-Efficient Distributed Learning via Compressed Momentum Filtering

2024-09-13 · Changxin Liu, Yanghao Li, Yuhao Yi, Karl H. Johansson

Distributed learning has become the standard approach for training large-scale machine learning models across private data silos. While distributed learning enhances privacy preservation and training efficiency, it faces…

BenchmarkingBinary Classificationimage-classificationImage Classification

Accelerating Byzantine-Robust Distributed Learning with Compressed Communication via Double Momentum and Variance Reduction

2026-03-16 · Yanghao Li, Changxin Liu, Yuhao Yi arxiv

In collaborative and distributed learning, Byzantine robustness reflects a major facet of optimization algorithms. Such distributed algorithms are often accompanied by transmitting a large number of parameters, so commun…

Variance Reduction is an Antidote to Byzantines: Better Rates, Weaker Assumptions and Communication Compression as a Cherry on the Top

2022-06-01 · Eduard Gorbunov, Samuel Horváth, Peter Richtárik, Gauthier Gidel

Byzantine-robustness has been gaining a lot of attention due to the growth of the interest in collaborative and federated learning. However, many fruitful directions, such as the usage of variance reduction for achieving…

Federated Learning

Byzantine Robustness and Partial Participation Can Be Achieved at Once: Just Clip Gradient Differences

2023-11-23 · Grigory Malinovsky, Peter Richtárik, Samuel Horváth, Eduard Gorbunov

Distributed learning has emerged as a leading paradigm for training large machine learning models. However, in real-world scenarios, participants may be unreliable or malicious, posing a significant challenge to the inte…