Research map / Optimization / Distributed and compressed
Variance Reduction is an Antidote to Byzantines: Better Rates, Weaker Assumptions and Communication Compression as a Cherry on the Top
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Use variance-reduced Byz-VR-MARINA instead of the SGD baseline | Distributed logistic regression with Byzantine workers · a9a, homogeneous clients, clean and adversarial attacks | convergence rate | Helped |
| Use Byz-VR-MARINA instead of compressed SGD or BR-DIANA | Distributed logistic regression with compressed communication · a9a, attacks other than bit flipping | optimality gap | Helped |
| Use Byz-VR-MARINA instead of compressed SGD or BR-DIANA | Distributed logistic regression with compressed communication · a9a, bit flipping attack | optimality gap | Mixed |
| Add RandK sparsification with K = 0.1 d (ablation) | Byz-VR-MARINA · a9a, ALIE attack | communication cost | Helped |
| Use importance sampling instead of uniform sampling for stochastic gradient differences (ablation) | Byz-VR-MARINA · Stochastic gradient sampling | communication-round complexity | Helped |
Closest papers by meaning
- Byzantine Robustness and Partial Participation Can Be Achieved at Once: Just Clip Gradient Differences (NeurIPS 2024 · similarity 0.92)
- From Inexact Gradients to Byzantine Robustness: Acceleration and Optimization under Similarity (NeurIPS 2026 · similarity 0.90)
- On the Tension between Byzantine Robustness and No-Attack Accuracy in Distributed Learning (ICML 2025 · similarity 0.89)
- Weight for Robustness: A Comprehensive Approach towards Optimal Fault-Tolerant Asynchronous ML (NeurIPS 2024 · similarity 0.88)
- Efficient Federated Learning against Byzantine Attacks and Data Heterogeneity via Aggregating Normalized Gradients (NeurIPS 2025 · similarity 0.88)
- Fault Tolerant ML: Efficient Meta-Aggregation and Synchronous Training (ICML 2024 · similarity 0.87)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.