Papers › Domain-independent Dominance of Adaptive Methods

Domain-independent Dominance of Adaptive Methods

4 Dec 2019CVPR 2021 1arXiv:1912.01823archive 2025-07-28

Pedro Savarese, David Mcallester, Sudarshan Babu, Michael Maire

From a simplified analysis of adaptive methods, we derive AvaGrad, a new optimizer which outperforms SGD on vision tasks when its adaptability is properly tuned. We observe that the power of our method is partially explained by a decoupling of learning rate and adaptability, greatly simplifying hyperparameter search. In light of this observation, we demonstrate that, against conventional wisdom, Adam can also outperform SGD on vision tasks, as long as the coupling between its learning rate and adaptability is taken into account. In practice, AvaGrad matches the best results, as measured by generalization accuracy, delivered by any existing optimizer (SGD or adaptive) across image classification (CIFAR, ImageNet) and character-level language modelling (Penn Treebank) tasks.

PaperPDFConference PDFCode

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

Code

lolemacs/avagrad officialmentioned in papermentioned on GitHubpytorch 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

Image ClassificationLanguage ModellingStochastic Optimizationimage-classification

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Stochastic Optimization CIFAR-10 WRN-28-10 - 200 Epochs Adam (eps-adjusted) Accuracy 96.36 #1 of 6 Archive leaderboard report
Stochastic Optimization CIFAR-10 WRN-28-10 - 200 Epochs AvaGrad Accuracy 96.2 #2 of 6 Archive leaderboard report
Stochastic Optimization CIFAR-10 WRN-28-10 - 200 Epochs SGD Accuracy 96.14 #3 of 6 Archive leaderboard report
Stochastic Optimization CIFAR-10 WRN-28-10 - 200 Epochs AdaShift Accuracy 95.92 #4 of 6 Archive leaderboard report
Stochastic Optimization CIFAR-10 WRN-28-10 - 200 Epochs AdamW Accuracy 95.89 #5 of 6 Archive leaderboard report
Stochastic Optimization CIFAR-10 WRN-28-10 - 200 Epochs AdaBound Accuracy 94.6 #6 of 6 Archive leaderboard report
Stochastic Optimization CIFAR-100 WRN-28-10 - 200 Epochs AvaGrad Accuracy 81.24 #1 of 6 Archive leaderboard report
Stochastic Optimization CIFAR-100 WRN-28-10 - 200 Epochs AdaShift Accuracy 81.12 #2 of 6 Archive leaderboard report
Stochastic Optimization CIFAR-100 WRN-28-10 - 200 Epochs Adam (eps-adjusted) Accuracy 81.04 #3 of 6 Archive leaderboard report
Stochastic Optimization CIFAR-100 WRN-28-10 - 200 Epochs SGD Accuracy 80.95 #4 of 6 Archive leaderboard report
Stochastic Optimization CIFAR-100 WRN-28-10 - 200 Epochs AdamW Accuracy 79.87 #5 of 6 Archive leaderboard report
Stochastic Optimization CIFAR-100 WRN-28-10 - 200 Epochs AdaBound Accuracy 77.24 #6 of 6 Archive leaderboard report
Stochastic Optimization ImageNet ResNet-50 - 90 Epochs AvaGrad Top 1 Accuracy 76.51 #1 of 4 Archive leaderboard report
Stochastic Optimization ImageNet ResNet-50 - 90 Epochs SGD Top 1 Accuracy 75.99 #2 of 4 Archive leaderboard report
Stochastic Optimization ImageNet ResNet-50 - 90 Epochs AdamW Top 1 Accuracy 72.9 #3 of 4 Archive leaderboard report
Stochastic Optimization ImageNet ResNet-50 - 90 Epochs AdaBound Top 1 Accuracy 72.01 #4 of 4 Archive leaderboard report
Stochastic Optimization Penn Treebank (Character Level) 3x1000 LSTM - 500 Epochs AvaGrad Bit per Character (BPC) 1.175 #1 of 4 Archive leaderboard report
Stochastic Optimization Penn Treebank (Character Level) 3x1000 LSTM - 500 Epochs AdamW Bit per Character (BPC) 1.23 #2 of 4 Archive leaderboard report
Stochastic Optimization Penn Treebank (Character Level) 3x1000 LSTM - 500 Epochs AdaShift Bit per Character (BPC) 1.274 #3 of 4 Archive leaderboard report
Stochastic Optimization Penn Treebank (Character Level) 3x1000 LSTM - 500 Epochs AdaBound Bit per Character (BPC) 2.863 #4 of 4 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Methods

AdamSGD

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