Papers › SGDR: Stochastic Gradient Descent with Warm Restarts

SGDR: Stochastic Gradient Descent with Warm Restarts

13 Aug 2016arXiv:1608.03983archive 2025-07-28

Ilya Loshchilov, Frank Hutter

Restart techniques are common in gradient-free optimization to deal with multimodal functions. Partial warm restarts are also gaining popularity in gradient-based optimization to improve the rate of convergence in accelerated gradient schemes to deal with ill-conditioned functions. In this paper, we propose a simple warm restart technique for stochastic gradient descent to improve its anytime performance when training deep neural networks. We empirically study its performance on the CIFAR-10 and CIFAR-100 datasets, where we demonstrate new state-of-the-art results at 3.14% and 16.21%, respectively. We also demonstrate its advantages on a dataset of EEG recordings and on a downsampled version of the ImageNet dataset. Our source code is available at https://github.com/loshchil/SGDR

PaperPDFCodeCode Syntology ranMethod as code

In Syntology View this paper on Syntology, its page in Syntology's graph. That page adds which of its functions are also present in other papers' code Syntology holds, with dates, and its dated run record as JSON and a README badge. It also 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

The archive's official repository (loshchil/SGDR) was not harvested; the 17 samples below come from 10 community repositories.

Syntology 17 code samples harvested from 10 repositories linked to this paper: 16 ran (of which 7 constructed an object rather than computing a result; 14 with no instrument failure: 0 honoured, 0 violated, 14 with no contract checked; 2 where Syntology's instrument failed) · 1 unverified. Where Syntology's instrument failed: 2 where our contract draft was wrong; failures of Syntology's instrument, not of the code.

By repository: community (archive-listed): 17 samples from 10 repositories, 16 ran (14 with no instrument failure; 2 where Syntology's instrument failed). The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

No sample is linked to this paper from 4 of its repositories that Syntology harvested for other papers:

Levels written between 2026-09-22 and 2026-09-26 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed. 1 of the 16 samples that ran has no recorded date.

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

loshchil/SGDR officialmentioned in papermentioned on GitHubSyntology: not harvested report
Gjiangtao/A-Deep-Supervised-Edge-Optimization-Algorithm-for-Salt-Body-Segmentation mentioned on GitHubpytorchSyntology: 1 sample, 1 ran (0 instrument-failed) · 0 unverified report
Harshvardhan1/cyclic-learning-schedulers-pytorch mentioned on GitHubpytorchSyntology: 1 sample, 1 ran (0 instrument-failed) · 0 unverified report
Jmak12/Iris1 mentioned on GitHubpytorchSyntology: 1 sample, 1 ran (0 instrument-failed) · 0 unverified report
MrtnMndt/Rethinking_CNN_Layerwise_Feature_Amounts mentioned on GitHubpytorchSyntology: 1 sample, 1 ran (0 instrument-failed) · 0 unverified report
PatriciaRodrigues1994/pytorch-facebook-udacity-challenge mentioned on GitHubpytorchSyntology: not harvested report
SevenZhan/Pytorch mentioned on GitHubpytorchSyntology: 1 sample, 1 ran (0 instrument-failed) · 0 unverified report
abhuse/cyclic-cosine-decay mentioned on GitHubpytorchSyntology: 1 sample, 1 ran (0 instrument-failed) · 0 unverified report
bluesky314/Cyclical_LR_Scheduler_With_Decay_Pytorch mentioned on GitHubpytorchSyntology: 1 sample, 1 ran (0 instrument-failed) · 0 unverified report
buptlwz/mabp mentioned on GitHubpytorchMITSyntology: 7 samples, 7 ran (2 instrument-failed) · 0 unverified report
gurucharanmk/PyTorch_CosineAnnealingWithRestartsLR mentioned on GitHubpytorchSyntology: 1 sample, 1 ran (0 instrument-failed) · 0 unverified report
jae1213/CyclicalLR_SGDR mentioned on GitHubpytorchSyntology: 2 samples, 1 ran (0 instrument-failed) · 1 unverified report
jolibrain/caffe mentioned on GitHubSyntology: not harvested report
longjj/Caffe-SGDR mentioned on GitHubSyntology: 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
tmbluth/Distracted-Drivers-Detection mentioned on GitHubSyntology: not harvested report
vinuni-vishc/transformer-gait-analysis mentioned on GitHubpytorchSyntology: not harvested report
yumatsuoka/check_cosine_annealing_lr mentioned on GitHubtorchSyntology: not harvested report
1hb6s7t/SGDR-ms mindsporeSyntology: not harvested report
MindCode-4/code-13 mindsporeSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Randomized Quantization: A Generic Augmentation for Data Agnostic Self-supervised Learning (arXiv:2212.08663; that paper's own run record: 4 ran (of which 0 constructed an object rather than computing a result; 4 with no instrument failure: 0 honoured, 0 violated, 4 with no contract checked; 0 where Syntology's instrument failed) · 1 unverified); nothing checks that they implement this paper's method. report
MindCode-4/code-6 mindsporeSyntology: not harvested report
MindCode-4/code-9 mindsporeApache-2.0Syntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to SA-MLP: Distilling Graph Knowledge from GNNs into Structure-Aware MLP (arXiv:2210.09609; that paper's own run record: 4 ran (of which 0 constructed an object rather than computing a result; 4 with no instrument failure: 0 honoured, 0 violated, 4 with no contract checked; 0 where Syntology's instrument failed) · 1 unverified); nothing checks that they implement this paper's method. report
pwc-1/Paper-9 mindsporeSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to RandAugment: Practical automated data augmentation with a reduced search space (arXiv:1909.13719; that paper's own run record: 58 ran (of which 1 constructed an object rather than computing a result; 7 with no instrument failure: 0 honoured, 1 violated, 6 with no contract checked; 51 where Syntology's instrument failed) · 7 unverified); nothing checks that they implement this paper's method. 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

17 samples harvested: 16 ran (of which 7 constructed an object rather than computing a result; 14 with no instrument failure: 0 honoured, 0 violated, 14 with no contract checked; 2 where Syntology's instrument failed) · 1 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Syntology's record for this page last changed ; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

2ran · our draft was wrong
14ran
1unverified

Licence: 7 of the 17 samples are pointer only, meaning Syntology does not serve that copy's text. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from 10 repositories linked to this paper, official or community; the list below is grouped by repository, each group headed by its name and what it is to the paper. “Ran” means the sample executed on a synthesized input. It does not mean the output is correct, and nothing here reproduces the paper's results. A line beginning “fixture:” under a sample that ran gives the arguments Syntology's fixture called the sample with, each as name:type/rank/element kind (? where none was recorded), and the type of what the fixture's run returned, exactly as the harness recorded them; the arguments and the returned type were recorded by two separate runs of the same fixture. No output value is shown, and a returned type says nothing about whether the output was right. “Level recorded” gives the UTC date when Syntology's graph wrote the sample's verification level, not when the code executed; where the level was written before Syntology's graph kept that date, the row says “level recorded, date not recorded”. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself. “Our draft was wrong” and “the contract check could not run” are failures of Syntology's instrument, not of the code: the sample ran on Syntology's synthesized input, and then the drafted contract did not describe it, or the check itself raised an error (RAISES), lacked a dependency (DEP_MISSING) or ran out of time (TIMEOUT); hover the label for which. A metamorphic tier, where shown, is evidence of the output's shape, not that the sample implements the method.

Each sample ends with its code_sha256. A sample's code_sha256 is Syntology's identity for that exact code. It is printed in full (16 hexadecimal characters for every sample in this build), and get_code(code_sha256="…") takes it exactly as printed. An agent calls it on Syntology's MCP service (how to connect); click an identity to copy that call.

Repository labels, per sample. official repository: The archive marks this repository official for the paper. named in the paper: The archive records that the paper mentions this repository; it is not marked official. community (archive-listed): In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper. found in paper text by Syntology: Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted. community: Not in the archive's code links for this paper; a community repository Syntology harvested. Samples are grouped by repository: a repository marked official first, then by how directly the repository is tied to the paper, then in the archive's order; never by how many ran. Licence labels name the repository's licence as recorded at harvest. “Pointer only” means Syntology does not serve that copy's text, for one of six reasons: no licence file was found; the licence was not identified; the licence is recorded as permissive but that copy's record is not marked cleared; the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar); this paper's copy was not located, so its licence is not recorded; or this paper's copy is cleared, but the stored copy of the same code that Syntology serves is not. Some licences outside that list permit redistribution, such as WTFPL, and GPL-3.0 under its conditions; they are simply not on the list. Hover a licence label for the reason. File links open the file on GitHub at the default branch, which may have changed since the harvest.

A function or class name links to its page when the name occurs in the harvested code of at least 5 papers; the name is a string, not an identity. A census tag appears when Syntology's census ran implementations with that name on one shared input per bucket: it gives how many were compared, from how many papers, and how many distinct outputs they produced, and names this sample's bucket when this sample was one of them.

Gjiangtao/A-Deep-Supervised-Edge-Optimization-Algorithm-for-Salt-Body-Segmentation community (archive-listed) 1 sample, 1 ran (0 instrument-failed) · 0 unverified
CosineAnnealingLR_with_Restart config/res34_loss_naive_nodice_hp_scse_ds_border/loss/cyclic_lr.py ran no licence file found · pointer only · 4f7499d2bb391700 · report fixture: returned CosineAnnealingLR_with_Restart · level recorded 2026-09-22
Harshvardhan1/cyclic-learning-schedulers-pytorch community (archive-listed) 1 sample, 1 ran (0 instrument-failed) · 0 unverified
CyclicCosAnnealingLR cyclicLR.py ran · metamorphic tier: well formed no licence file found · pointer only · 8a9fabc6ccb89932 · report fixture: returned list · level recorded 2026-09-22
Jmak12/Iris1 community (archive-listed) 1 sample, 1 ran (0 instrument-failed) · 0 unverified
PolyLR optimizers/lr_scheduler.py ran no licence file found · pointer only · 899d6ddaba2af2a7 · report fixture: returned PolyLR · level recorded 2026-09-22
MrtnMndt/Rethinking_CNN_Layerwise_Feature_Amounts community (archive-listed) 1 sample, 1 ran (0 instrument-failed) · 0 unverified
LearningRateScheduler lib/Training/learning_rate_scheduling.py ran BSD-3-Clause (permissive) · e2d73df88181d0fb · report fixture: returned LearningRateScheduler · level recorded 2026-09-22
SevenZhan/Pytorch community (archive-listed) 1 sample, 1 ran (0 instrument-failed) · 0 unverified
CosineLR schedulers.py ran no licence file found · pointer only · fe5e79b70f387bec · report fixture: returned CosineLR · level recorded 2026-09-22
abhuse/cyclic-cosine-decay community (archive-listed) 1 sample, 1 ran (0 instrument-failed) · 0 unverified
CyclicCosineDecayLR scheduler.py ran MIT (permissive) · 926fab0aa32bd97a · report fixture: returned CyclicCosineDecayLR · level recorded 2026-09-22
bluesky314/Cyclical_LR_Scheduler_With_Decay_Pytorch community (archive-listed) 1 sample, 1 ran (0 instrument-failed) · 0 unverified
CyclicCosAnnealingLR cyclicLR.py ran no licence file found · pointer only · 43331e0456fed368 · report fixture: signature epoch:NoneType/0/? · returned NoneType · level recorded 2026-09-24
buptlwz/mabp community (archive-listed) 7 samples, 7 ran (2 instrument-failed) · 0 unverified
computeIoU in 11 papers gres_model/evaluation/refer_evaluation.py ran MIT (permissive) · 745bf09fefeb50c3 · report fixture: signature pred_seg:ndarray/2/bool, gd_seg:ndarray/2/bool · returned tuple · level recorded 2026-09-26
nested_tensor_from_tensor_list in 60 papers gres_model/utils/misc.py ran · our draft was wrong MIT (permissive) · 58cc9ff3bf75e753 · report fixture: signature tensor_list:list/1/Tensor · returned NestedTensor · level recorded 2026-09-25
prepare_data in 87 papers gres_model/data_prepare.py ran MIT (permissive) · 755b208f128fbce6 · report fixture: signature data:list/1/dict, train:bool/0/bool · returned dict · level recorded 2026-09-26
refer_bce_loss gres_model/modeling/criterion-so.py ran MIT (permissive) · 67dae4afd1febb5e · report fixture: signature inputs:Tensor/2/float, targets:Tensor/2/float, weight:NoneType/0/? · returned Tensor · level recorded 2026-09-26
refer_ce_loss gres_model/modeling/criterion-so.py ran MIT (permissive) · 7b44eda6b63c9a3a · report fixture: signature inputs:Tensor/2/float, targets:Tensor/1/int, weight:Tensor/1/float · returned Tensor · level recorded 2026-09-26
window_partition in 377 papers
census: 13 implementations from 17 papers, 5 distinct outputs across 3 buckets, one shared input per bucket
gres_model/modeling/criterion-so.py ran fingerprinted MIT (permissive) · 2f4f1157ededdc6b · report fixture: signature x:Tensor/4/float, window_size:int/0/int · returned Tensor · level recorded 2026-09-26
window_reverse in 296 papers gres_model/modeling/backbone/swin.py ran · our draft was wrong MIT (permissive) · 61bf152e6a42a184 · report fixture: signature windows:Tensor/4/float, window_size:int/0/int, H:int/0/int, W:int/0/int · returned Tensor · level recorded, date not recorded
gurucharanmk/PyTorch_CosineAnnealingWithRestartsLR community (archive-listed) 1 sample, 1 ran (0 instrument-failed) · 0 unverified
CosineAnnealingWithRestartsLR CosineAnnealingWithRestartsLR.py ran MIT (permissive) · f4a8879aa783ad15 · report fixture: returned CosineAnnealingWithRestartsLR · level recorded 2026-09-22
jae1213/CyclicalLR_SGDR community (archive-listed) 2 samples, 1 ran (0 instrument-failed) · 1 unverified
SGDRLR LRscheduler.py ran no licence file found · pointer only · 6e9ad0df07eed6da · report fixture: returned SGDRLR · level recorded 2026-09-22
CustomLR LRscheduler.py unverified no licence file found · pointer only · 803f499019d43ed0 · report

Method as code Syntology

What Syntology's graph records as this paper's claimed method, lifted into code: each assembly names its bricks (functions and classes), each with its own role and its own verification level. An assembly is not a check that the code implements the method correctly or reproduces the paper's results. Levels are each brick's own verification level in Syntology's graph (level 0: no recorded run), listed brick by brick and never combined into one level; “ran” means the brick executed on a synthesized input, not that its output is correct, and the instrument figure counts failures of Syntology's instrument, not of the code. Where a model read the code against the paper's claim, each assembly's label verdict is one model's judgement, kept in this page's JSON twin; it is not calibrated, so no card prints it. Siblings are assemblies lifted from this paper in the same batch for the same method: alternative readings kept side by side; none replaces or supersedes another.

9 assemblies on this paper: 9 harvested, 0 generated, listed in 1 groups (siblings lifted in one batch form one group).

9 sibling assemblies for SGDR: STOCHASTIC GRADIENT DESCENT WITH WARM RESTARTS (harvested), lifted in the same batch on 2026-09-09: alternative readings kept side by side; none replaces or supersedes another.

harvested Assembly for SGDR: STOCHASTIC GRADIENT DESCENT WITH WARM RESTARTS (graph method: SGDR: STOCHASTIC GRADIENT DESCENT WITH WARM RESTARTS). 1 brick; individually on a fixture: 1 ran (of which 1 constructed an object rather than computing a result; 1 with no instrument failure: 0 honoured, 0 violated, 1 with no contract checked; 0 where Syntology's instrument failed) · 0 unverified; the one sample that ran constructed an object rather than computing a result. The assembly has not been run end-to-end.

one of 9 siblings

  1. PolyLR entry · level 1 · ran · 899d6ddaba2af2a7

harvested Assembly for SGDR: STOCHASTIC GRADIENT DESCENT WITH WARM RESTARTS (graph method: SGDR: STOCHASTIC GRADIENT DESCENT WITH WARM RESTARTS). 1 brick; individually on a fixture: 1 ran (of which 1 constructed an object rather than computing a result; 1 with no instrument failure: 0 honoured, 0 violated, 1 with no contract checked; 0 where Syntology's instrument failed) · 0 unverified; the one sample that ran constructed an object rather than computing a result. The assembly has not been run end-to-end.

one of 9 siblings

  1. CosineLR entry · level 1 · ran · fe5e79b70f387bec

harvested Assembly for SGDR: STOCHASTIC GRADIENT DESCENT WITH WARM RESTARTS (graph method: SGDR: STOCHASTIC GRADIENT DESCENT WITH WARM RESTARTS). 1 brick; individually on a fixture: 1 ran (of which 1 constructed an object rather than computing a result; 1 with no instrument failure: 0 honoured, 0 violated, 1 with no contract checked; 0 where Syntology's instrument failed) · 0 unverified; the one sample that ran constructed an object rather than computing a result. The assembly has not been run end-to-end.

one of 9 siblings

  1. CosineAnnealingLR_with_Restart entry · level 1 · ran · 4f7499d2bb391700

harvested Assembly for SGDR: STOCHASTIC GRADIENT DESCENT WITH WARM RESTARTS (graph method: SGDR: STOCHASTIC GRADIENT DESCENT WITH WARM RESTARTS). 1 brick; individually on a fixture: 1 ran (of which 1 constructed an object rather than computing a result; 1 with no instrument failure: 0 honoured, 0 violated, 1 with no contract checked; 0 where Syntology's instrument failed) · 0 unverified; the one sample that ran constructed an object rather than computing a result. The assembly has not been run end-to-end.

one of 9 siblings

  1. CosineAnnealingWithRestartsLR entry · level 1 · ran · f4a8879aa783ad15

harvested Assembly for SGDR: STOCHASTIC GRADIENT DESCENT WITH WARM RESTARTS (graph method: SGDR: STOCHASTIC GRADIENT DESCENT WITH WARM RESTARTS). 1 brick; individually on a fixture: 1 ran (of which 1 constructed an object rather than computing a result; 1 with no instrument failure: 0 honoured, 0 violated, 1 with no contract checked; 0 where Syntology's instrument failed) · 0 unverified; the one sample that ran constructed an object rather than computing a result. The assembly has not been run end-to-end.

one of 9 siblings

  1. CyclicCosineDecayLR entry · level 1 · ran · 926fab0aa32bd97a

harvested Assembly for SGDR: STOCHASTIC GRADIENT DESCENT WITH WARM RESTARTS (graph method: SGDR: STOCHASTIC GRADIENT DESCENT WITH WARM RESTARTS). 2 bricks; individually on a fixture: 1 ran (of which 1 constructed an object rather than computing a result; 1 with no instrument failure: 0 honoured, 0 violated, 1 with no contract checked; 0 where Syntology's instrument failed) · 1 unverified; the one sample that ran constructed an object rather than computing a result. The assembly has not been run end-to-end.

one of 9 siblings

  1. SGDRLR entry · level 1 · ran · 6e9ad0df07eed6da
  2. CustomLR helper · level 0 · unverified · 803f499019d43ed0

harvested Assembly for SGDR: STOCHASTIC GRADIENT DESCENT WITH WARM RESTARTS (graph method: SGDR: STOCHASTIC GRADIENT DESCENT WITH WARM RESTARTS). 1 brick; individually on a fixture: 1 ran (of which 1 constructed an object rather than computing a result; 1 with no instrument failure: 0 honoured, 0 violated, 1 with no contract checked; 0 where Syntology's instrument failed) · 0 unverified; the one sample that ran constructed an object rather than computing a result. The assembly has not been run end-to-end.

one of 9 siblings

  1. LearningRateScheduler entry · level 1 · ran · e2d73df88181d0fb

harvested Assembly for SGDR: STOCHASTIC GRADIENT DESCENT WITH WARM RESTARTS (graph method: SGDR: STOCHASTIC GRADIENT DESCENT WITH WARM RESTARTS). 1 brick; individually on a fixture: 1 ran (of which 0 constructed an object rather than computing a result; 1 with no instrument failure: 0 honoured, 0 violated, 1 with no contract checked; 0 where Syntology's instrument failed) · 0 unverified. The assembly has not been run end-to-end.

one of 9 siblings

  1. CyclicCosAnnealingLR entry · level 1 · ran · 8a9fabc6ccb89932

harvested Assembly for SGDR: STOCHASTIC GRADIENT DESCENT WITH WARM RESTARTS (graph method: SGDR: STOCHASTIC GRADIENT DESCENT WITH WARM RESTARTS). 1 brick; individually on a fixture: 1 ran (of which 0 constructed an object rather than computing a result; 1 with no instrument failure: 0 honoured, 0 violated, 1 with no contract checked; 0 where Syntology's instrument failed) · 0 unverified. The assembly has not been run end-to-end.

one of 9 siblings

  1. CyclicCosAnnealingLR entry · level 1 · ran · 43331e0456fed368

Each brick ends with its code_sha256, Syntology's identity for that exact code; click one to copy the get_code call that returns it. For agents, on Syntology's MCP service: assembly(arxiv_id="1608.03983") returns these assemblies with each brick's role and wiring (how to connect).

Tasks

EEGElectroencephalogram (EEG)Stochastic Optimization

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Methods

Introduced by this paper: Cosine Annealing

Cosine Annealing

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