Research map / Optimization / Distributed and compressed
The Hidden Vulnerability of Distributed Learning in Byzantium
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 |
|---|---|---|---|
| apply the proposed attack to Krum (norm 2) | Krum · CIFAR-10 | accuracy: 33% | Hurt |
| apply the proposed attack to GeoMed (norm 2) | GeoMed · CIFAR-10 | model usefulness | Hurt |
| apply the attack for the first 50 epochs, then stop | Krum and GeoMed · MNIST | model effectiveness | Hurt |
| use Bulyan with Krum as its base rule instead of Krum alone | Krum · MNIST | accuracy under attack | Helped |
| use Bulyan with Krum as its base rule instead of Krum alone | Krum · CIFAR-10 | accuracy under attack | Helped |
| use Bulyan with Krum as its base rule instead of averaging (ablation) | distributed SGD without Byzantine workers · MNIST | accuracy at epoch 250: 24 images/batch | Mixed |
| use Bulyan with Krum as its base rule instead of averaging (ablation) | distributed SGD without Byzantine workers · CIFAR-10 | accuracy at epoch 350: 36 | Mixed |
| increase the learning rate | the proposed Byzantine attack · MNIST and CIFAR-10 | attack effectiveness | Helped |
| lower the batch size | the proposed Byzantine attack · MNIST and CIFAR-10 | attack effectiveness | Helped |
Closest papers by meaning
- Distributed Momentum for Byzantine-resilient Stochastic Gradient Descent (ICLR 2021 · similarity 0.90)
- Machine Learning with Adversaries: Byzantine Tolerant Gradient Descent (NeurIPS 2017 · similarity 0.89)
- From Inexact Gradients to Byzantine Robustness: Acceleration and Optimization under Similarity (NeurIPS 2026 · similarity 0.86)
- Byzantine Robustness and Partial Participation Can Be Achieved at Once: Just Clip Gradient Differences (NeurIPS 2024 · similarity 0.85)
- On the Tension between Byzantine Robustness and No-Attack Accuracy in Distributed Learning (ICML 2025 · similarity 0.84)
- Fault Tolerant ML: Efficient Meta-Aggregation and Synchronous Training (ICML 2024 · similarity 0.83)
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.