Papers › Lookahead Optimizer: k steps forward, 1 step back

Lookahead Optimizer: k steps forward, 1 step back

19 Jul 2019NeurIPS 2019 12arXiv:1907.08610archive 2025-07-28

Michael R. Zhang, James Lucas, Geoffrey Hinton, Jimmy Ba

The vast majority of successful deep neural networks are trained using variants of stochastic gradient descent (SGD) algorithms. Recent attempts to improve SGD can be broadly categorized into two approaches: (1) adaptive learning rate schemes, such as AdaGrad and Adam, and (2) accelerated schemes, such as heavy-ball and Nesterov momentum. In this paper, we propose a new optimization algorithm, Lookahead, that is orthogonal to these previous approaches and iteratively updates two sets of weights. Intuitively, the algorithm chooses a search direction by looking ahead at the sequence of fast weights generated by another optimizer. We show that Lookahead improves the learning stability and lowers the variance of its inner optimizer with negligible computation and memory cost. We empirically demonstrate Lookahead can significantly improve the performance of SGD and Adam, even with their default hyperparameter settings on ImageNet, CIFAR-10/100, neural machine translation, and Penn Treebank.

PaperPDFConference PDFCode

In Syntology View this paper on Syntology, its page in Syntology's graph. That page lists the repositories linked to the paper, the abstract and the calls for agents.

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

For agents, on Syntology's MCP service (how to connect):

Code

19 repositories listed; official and paper-mentioned ones first.

michaelrzhang/lookahead officialmentioned in papermentioned on GitHubpytorchMITSyntology: not harvested report
201419/Optimizer-PyTorch mentioned on GitHubpytorchApache-2.0Syntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to On the Variance of the Adaptive Learning Rate and Beyond (arXiv:1908.03265; that paper's own run record: 12 ran (of which 0 constructed an object rather than computing a result; 11 with no instrument failure: 2 honoured, 0 violated, 9 with no contract checked; 1 where Syntology's instrument failed) · 1 unverified); nothing checks that they implement this paper's method. report
Abhimanyu08/Lookahead_Optimizer mentioned on GitHubpytorchSyntology: not harvested report
COMP6248-Reproducability-Challenge/LookaheadOptimizer mentioned on GitHubpytorchSyntology: not harvested report
DreamInvoker/LookaheadOptimizer-mx-implementation mentioned on GitHubmxnetSyntology: not harvested report
HamadYA/GhostFaceNets mentioned on GitHubtfMITSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Gradient Centralization: A New Optimization Technique for Deep Neural Networks (arXiv:2004.01461; that paper's own run record: 18 ran (of which 0 constructed an object rather than computing a result; 17 with no instrument failure: 1 honoured, 0 violated, 16 with no contract checked; 1 where Syntology's instrument failed) · 11 unverified); nothing checks that they implement this paper's method. report
SdahlSean/RangerOptimizerTensorflow mentioned on GitHubtfSyntology: not harvested report
allen108108/Model-Optimizer_Implementation mentioned on GitHubSyntology: not harvested report
alphadl/lookahead.pytorch mentioned on GitHubpytorchMITSyntology: not harvested report
bojone/keras_lookahead mentioned on GitHubSyntology: not harvested report
chizhu/BDC2019 mentioned on GitHubtfSyntology: not harvested report
kpe/params-flow mentioned on GitHubtfMITSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to On the Variance of the Adaptive Learning Rate and Beyond (arXiv:1908.03265; that paper's own run record: 12 ran (of which 0 constructed an object rather than computing a result; 11 with no instrument failure: 2 honoured, 0 violated, 9 with no contract checked; 1 where Syntology's instrument failed) · 1 unverified); nothing checks that they implement this paper's method. report
mnikitin/LookaheadOptimizer-mx mentioned on GitHubmxnetSyntology: not harvested report
nachiket273/lookahead_pytorch mentioned on GitHubpytorchSyntology: not harvested report
nsarang/lookahead_keras mentioned on GitHubtfMITSyntology: not harvested report
rwightman/pytorch-image-models mentioned on GitHubpytorchSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to AdamP: Slowing Down the Slowdown for Momentum Optimizers on Scale-invariant Weights (arXiv:2006.08217; that paper's own run record: 4 ran (of which 1 constructed an object rather than computing a result; 2 with no instrument failure: 1 honoured, 0 violated, 1 with no contract checked; 2 where Syntology's instrument failed) · 0 unverified); nothing checks that they implement this paper's method. report
wkcn/LookaheadOptimizer-mx mentioned on GitHubmxnetSyntology: not harvested report
zhangtj1996/lookahead-sgd-adam-rmsprop- mentioned on GitHubtfSyntology: not harvested report
dseuss/pytorch-lookahead-optimizer pytorchSyntology: not harvested 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.

A paper named beside a repository with no sample linked to this paper is shown with that paper's own run record, not this paper's: “ran” means executed on a synthesized input, not that the code is correct, and “where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Papers are listed in arXiv-id order, at most three per repository.

Code Syntology ran Syntology

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

Tasks

Image ClassificationMachine TranslationStochastic OptimizationTranslation

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Stochastic Optimization CIFAR-10 ResNet-18 - 200 Epochs Lookahead Accuracy 95.27 #2 of 4 Archive leaderboard report
Stochastic Optimization CIFAR-10 ResNet-18 - 200 Epochs SGD Accuracy 95.23 #3 of 4 Archive leaderboard report
Stochastic Optimization CIFAR-10 ResNet-18 - 200 Epochs ADAM Accuracy 94.84 #4 of 4 Archive leaderboard report
Stochastic Optimization ImageNet ResNet-50 - 50 Epochs Lookahead Top 1 Accuracy 75.13% #1 of 2 Archive leaderboard report
Stochastic Optimization ImageNet ResNet-50 - 50 Epochs SGD Top 5 Accuracy 92.15% #2 of 2 Archive leaderboard report
Stochastic Optimization ImageNet ResNet-50 - 60 Epochs Lookahead Top 1 Accuracy 75.49% #1 of 2 Archive leaderboard report
Stochastic Optimization ImageNet ResNet-50 - 60 Epochs Lookahead Top 5 Accuracy 92.53 #1 of 2 Archive leaderboard report
Stochastic Optimization ImageNet ResNet-50 - 60 Epochs SGD Top 1 Accuracy 75.15% #2 of 2 Archive leaderboard report
Stochastic Optimization ImageNet ResNet-50 - 60 Epochs SGD Top 5 Accuracy 92.56 #2 of 2 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

1x1 ConvolutionAdamAverage PoolingBatch NormalizationBottleneck Residual BlockConvolutionGlobal Average PoolingKaiming InitializationLookaheadMax PoolingReLUResidual BlockResidual Connection

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