Pith. sign in

REVIEW 1 cited by

Taming Momentum in a Distributed Asynchronous Environment

Not yet reviewed by Pith; the record is open.

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 1907.11612 v3 pith:FUXYSSH6 submitted 2019-07-26 cs.LG cs.DCstat.ML

classification cs.LGcs.DCstat.ML
keywords asynchronousgradientdistributedmomentumtrainingaccuracyconvergencefinal
verification ladder T0 review T1 audit T2 compute T3 formal

Signed reviews

No signed human review yet.

0 comments
read the original abstract

Although distributed computing can significantly reduce the training time of deep neural networks, scaling the training process while maintaining high efficiency and final accuracy is challenging. Distributed asynchronous training enjoys near-linear speedup, but asynchrony causes gradient staleness - the main difficulty in scaling stochastic gradient descent to large clusters. Momentum, which is often used to accelerate convergence and escape local minima, exacerbates the gradient staleness, thereby hindering convergence. We propose DANA: a novel technique for asynchronous distributed SGD with momentum that mitigates gradient staleness by computing the gradient on an estimated future position of the model's parameters. Thereby, we show for the first time that momentum can be fully incorporated in asynchronous training with almost no ramifications to final accuracy. Our evaluation on the CIFAR and ImageNet datasets shows that DANA outperforms existing methods, in both final accuracy and convergence speed while scaling up to a total batch size of 16K on 64 asynchronous workers.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

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

  1. Nesterov Method for Asynchronous Pipeline Parallel Optimization

    cs.LG 2025-05 conditional novelty 5.0 of 10

    A discounted Nesterov look-ahead update, implemented with NAdam, is claimed to converge sublinearly under fixed gradient delay and to beat synchronous pipeline baselines in language modeling.

Pith tools