Research map / Optimization / Distributed and compressed
Machine Learning with Adversaries: Byzantine Tolerant Gradient Descent
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 |
|---|---|---|---|
| Replace averaging with Krum aggregation (no Byzantine workers) | distributed SGD, spambase MLP · spambase, 0% Byzantine, batch size 3 | cross-validation error | Hurt |
| Krum aggregation under 33% Gaussian Byzantine workers vs averaging | distributed SGD, spambase MLP · spambase, 33% Gaussian Byzantine, batch size 3 | cross-validation error | Helped |
| Multi-Krum (m=n-f) aggregation under 33% Gaussian Byzantine workers (ablation) | distributed SGD, spambase MLP · spambase, 33% Gaussian Byzantine, batch size 3 | convergence speed vs averaging with 0% Byzantine | Helped |
| Increase mini-batch size from ~1 to around 20-30 with Krum (ablation) | Krum under 45% omniscient Byzantine workers · spambase and MNIST | error at round 500 | Helped |
Closest papers by meaning
- From Inexact Gradients to Byzantine Robustness: Acceleration and Optimization under Similarity (NeurIPS 2026 · similarity 0.92)
- Fault Tolerant ML: Efficient Meta-Aggregation and Synchronous Training (ICML 2024 · similarity 0.91)
- On the Tension between Byzantine Robustness and No-Attack Accuracy in Distributed Learning (ICML 2025 · similarity 0.90)
- Weight for Robustness: A Comprehensive Approach towards Optimal Fault-Tolerant Asynchronous ML (NeurIPS 2024 · similarity 0.90)
- Byzantine Robustness and Partial Participation Can Be Achieved at Once: Just Clip Gradient Differences (NeurIPS 2024 · similarity 0.90)
- Near-Optimal Resilient Aggregation Rules for Distributed Learning Using 1-Center and 1-Mean Clustering with Outliers (AAAI 2024 · similarity 0.89)
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.