Pith. sign in

REVIEW 2 cited by

Gossip training for deep learning

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 1611.09726 v1 pith:CVCNQVXF submitted 2016-11-29 cs.CV cs.LGstat.ML

Michael Blot , David Picard , Matthieu Cord , Nicolas Thome This is my paper · ORCID
classification cs.CVcs.LGstat.ML
keywords methodgossipgradientthreadstrainingadaptedaddressadvantage
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

We address the issue of speeding up the training of convolutional networks. Here we study a distributed method adapted to stochastic gradient descent (SGD). The parallel optimization setup uses several threads, each applying individual gradient descents on a local variable. We propose a new way to share information between different threads inspired by gossip algorithms and showing good consensus convergence properties. Our method called GoSGD has the advantage to be fully asynchronous and decentralized. We compared our method to the recent EASGD in \cite{elastic} on CIFAR-10 show encouraging results.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 2 Pith papers

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

  1. Factored Gossip DiLoCo: Reducing Blocking Communication in DiLoCo

    cs.LG 2026-06 unverdicted novelty 5.0 of 10

    Factored Gossip DiLoCo relaxes exact outer synchronization in DiLoCo to approximate gossip-based mixing, enabling non-blocking steps and a tunable trade-off between compute utilization and stability on up to billion-p...

  2. Load-Aware Training Scheduling for Model Circulation-based Decentralized Federated Learning

    cs.LG 2025-06 conditional novelty 5.0 of 10

    Load-aware Tram-FL greedily schedules the training node and per-label data amount each round to maximize samples trained per second, and simulations report faster convergence than three baselines.

Pith tools