Pith. sign in

REVIEW 6 cited by

Byzantine-Robust Decentralized Learning via ClippedGossip

Not yet reviewed by Pith; the record is open.

Re-run · record.json Download PDF Read on arXiv ↗

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2202.01545 v2 pith:JPOGEJNP submitted 2022-02-03 cs.LG cs.DCmath.OCstat.ML

Lie He , Sai Praneeth Karimireddy , Martin Jaggi This is my paper
classification cs.LGcs.DCmath.OCstat.ML
keywords byzantine-robustclippedgossipdecentralizedconsensuslearningtrainingunderworkers
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
abstract

In this paper, we study the challenging task of Byzantine-robust decentralized training on arbitrary communication graphs. Unlike federated learning where workers communicate through a server, workers in the decentralized environment can only talk to their neighbors, making it harder to reach consensus and benefit from collaborative training. To address these issues, we propose a ClippedGossip algorithm for Byzantine-robust consensus and optimization, which is the first to provably converge to a $O(\delta_{\max}\zeta^2/\gamma^2)$ neighborhood of the stationary point for non-convex objectives under standard assumptions. Finally, we demonstrate the encouraging empirical performance of ClippedGossip under a large number of attacks.

Discussion (0). Sign in to comment.

Forward citations

Cited by 6 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Enhancing Parallelism in Decentralized Stochastic Convex Optimization

    cs.LG 2025-06 conditional novelty 7.0 of 10

    DAT-SGD improves the parallelism bound in decentralized stochastic convex optimization from O((ρ√N)^(1/2)) to O(ρ√N), matching centralized rates on dense graphs.

  2. RED-SEGA:Resilient Decentralized Stochastic Proximal Optimization with Gradient Sketching over Time-Varying Networks

    math.OC 2026-07 conditional novelty 6.0 of 10

    RED-SEGA achieves Byzantine-resilient linear convergence for non-decomposable SRM via gradient sketching and norm-penalized aggregation over time-varying networks.

  3. Delayed Momentum Aggregation: Communication-efficient Byzantine-robust Federated Learning with Partial Participation

    cs.LG 2025-09 conditional novelty 6.0 of 10

    D-Byz-SGDM aggregates cached momentum from non-sampled clients together with fresh momentum from sampled clients, preserving Byzantine robustness under partial participation and achieving an optimal O(cδζ²/p) stationa...

  4. DICE: Data Influence Cascade in Decentralized Learning

    cs.LG 2025-07 conditional novelty 6.0 of 10

    DICE defines and approximates multi-hop data influence in decentralized learning, showing that influence is shaped by data, topology, and loss curvature.

  5. Generalization Error Analysis for Attack-Free and Byzantine-Resilient Decentralized Learning with Data Heterogeneity

    cs.LG 2025-06 conditional novelty 6.0 of 10

    Decentralized SGD generalization error is bounded by O(init/(µNZ)) plus noise and heterogeneity terms, with a Byzantine-attack term that persists as sample size grows.

  6. Trust-Aware Topology Learning for Dynamic Decentralized Federated Learning under Adversaries

    cs.DC 2026-08 conditional novelty 5.0 of 10

    DMTT screens both model updates and topology claims with a Beta-trust model, and on two HAR datasets it is the only tested method that beats local-only learning under 10 to 80 percent Byzantine devices.

Pith tools