Papers › Adaptivity without Compromise: A Momentumized, Adaptive, Dual Averaged Gradient Method...

Adaptivity without Compromise: A Momentumized, Adaptive, Dual Averaged Gradient Method for Stochastic Optimization

26 Jan 2021arXiv:2101.11075archive 2025-07-28

Aaron Defazio, Samy Jelassi

We introduce MADGRAD, a novel optimization method in the family of AdaGrad adaptive gradient methods. MADGRAD shows excellent performance on deep learning optimization problems from multiple fields, including classification and image-to-image tasks in vision, and recurrent and bidirectionally-masked models in natural language processing. For each of these tasks, MADGRAD matches or outperforms both SGD and ADAM in test set performance, even on problems for which adaptive methods normally perform poorly.

PaperPDFCode

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

Code

facebookresearch/madgrad officialmentioned on GitHubpytorch report
DarshanDeshpande/tf-madgrad mentioned on GitHubtf report
epsilon-deltta/ssd_guillotine mentioned on GitHubpytorch report
mlverse/madgrad mentioned on GitHubpytorch report
sdatkinson/madgrad mentioned on GitHubjax report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Stochastic Optimization

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Methods

Introduced by this paper: MADGRAD

AdaGradAdamMADGRADSGD

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections