Papers › Identity Mappings in Deep Residual Networks

Identity Mappings in Deep Residual Networks

16 Mar 2016arXiv:1603.05027archive 2025-07-28

Kaiming He, Xiangyu Zhang, Shaoqing Ren, Jian Sun

Deep residual networks have emerged as a family of extremely deep architectures showing compelling accuracy and nice convergence behaviors. In this paper, we analyze the propagation formulations behind the residual building blocks, which suggest that the forward and backward signals can be directly propagated from one block to any other block, when using identity mappings as the skip connections and after-addition activation. A series of ablation experiments support the importance of these identity mappings. This motivates us to propose a new residual unit, which makes training easier and improves generalization. We report improved results using a 1001-layer ResNet on CIFAR-10 (4.62% error) and CIFAR-100, and a 200-layer ResNet on ImageNet. Code is available at: https://github.com/KaimingHe/resnet-1k-layers

PaperPDFCodeCode Syntology ran

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

For agents, Syntology's MCP tool lists every function and class Syntology harvested from this paper and whether it ran (how to connect): get_harvested_code_for_paper(arxiv_id="1603.05027")

Code

Syntology Ran 3 of 25 code samples harvested from 5 repositories linked to this paper; 22 have no recorded run. Of those that ran: 1 ran · honoured contract; 2 ran · our draft was wrong.

By repository: community (archive-listed): 25 samples from 5 repositories, 3 ran. The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

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

KaimingHe/resnet-1k-layers officialmentioned in papermentioned on GitHubtorch report
1M50RRY/resnet18-preact mentioned on GitHubMIT report
Deci-AI/super-gradients mentioned on GitHubpytorch report
Graylab/deepH3-distances-orientations mentioned on GitHubpytorch report
Halesu/4th-ML100Days mentioned on GitHubtf report
HubertTW/1st-DL-CVMarathon mentioned on GitHub report
IMvision12/keras-vision-models mentioned on GitHubpytorch report
P2333/Max-Mahalanobis-Training mentioned on GitHubtfApache-2.0 report
P2333/Mixup-Inference mentioned on GitHubpytorch report
TuSimple/resnet.mxnet mentioned on GitHubtfMIT report
Zi-Pan/ResnetAgePrediciton mentioned on GitHubpytorch report
bazilas/matconvnet-ResNet mentioned on GitHubNOASSERTION report
brain-bzh/MCNN mentioned on GitHubpytorchMIT report
breadboykid/ResnetAgePrediciton mentioned on GitHubpytorch report
edufonseca/icassp19 mentioned on GitHubtfMIT report
farrell236/ResNetAE mentioned on GitHubtfMIT report
google-research/diffstride mentioned on GitHubtfApache-2.0 report
horse007666/ResNet mentioned on GitHubtfnot reachable when probed 2026-09-18 — repositories for recent papers often appear after camera-ready report
huangleiBuaa/DecorrelatedBN mentioned on GitHubpytorchBSD-2-Clause report
hysts/pytorch_resnet_preact mentioned on GitHubpytorch report
iArunava/ResNet mentioned on GitHubpytorchBSD-3-Clause report
jofas/master_thesis mentioned on GitHubtfGPL-3.0 report
junhocho/SRGAN mentioned on GitHubMIT report
kaseris/ILSVRCPlus mentioned on GitHubnot reachable when probed 2026-09-18 — repositories for recent papers often appear after camera-ready report
marload/ConvNets-TensorFlow2 mentioned on GitHubtf report
mindspore-courses/MindSpore-classification mentioned on GitHubmindsporeMIT report
osmr/imgclsmob mentioned on GitHubmxnetMIT report
poloclub/jpeg-defense mentioned on GitHubtfMIT report
raghakot/keras-resnet mentioned on GitHubtfNOASSERTION report
rickyHong/JPEG-Defense-repl mentioned on GitHubtfMIT report
seansoleyman/cifar10-resnet mentioned on GitHubtf report
serjtroshin/pytorch-cifar-models mentioned on GitHubpytorch report
sitegui/ceci-nest-pas-un-chat mentioned on GitHubtfMIT report
smc-x/ms-resnetv2 mentioned on GitHubmindsporeApache-2.0 report
statsu1990/ReZero-Cifar100 mentioned on GitHubpytorch report
tensorflow/models mentioned on GitHubtf report
tensorflow/models mentioned on GitHubtf report
tensorflow/models mentioned on GitHubtf report
tensorpack/tensorpack mentioned on GitHubtf report
uclaml/Frank-Wolfe-AdvML mentioned on GitHubtfApache-2.0 report
umich-vl/DecorrelatedBN mentioned on GitHubpytorchBSD-2-Clause report
wenxinxu/resnet-in-tensorflow mentioned on GitHubtf report
wenxinxu/resnet_in_tensorflow mentioned on GitHubtfMIT report
yatharthagarwal/x_ray mentioned on GitHubmxnet report
zjZSTU/ResNet mentioned on GitHubpytorch report
JoegameZhou/ms-resnetv2 mindsporeApache-2.0 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

25 samples harvested; 3 ran; 1 honoured the contract we drafted; 22 have no recorded run. Read from Syntology's graph 2026-09-24; that is when this build read the record, not when the samples ran.

1ran · honoured contract
2ran · our draft was wrong
22unverified

Licence: 1 of the 25 samples is 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 5 repositories linked to this paper, official or community; each sample names its own and says which. “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. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself; “our draft was wrong” and “fixture could not drive it” are failures of Syntology's instrument, not of the code.

Each sample ends with its code_sha256, Syntology's identity for that exact code. An agent fetches the stored sample with Syntology's MCP tool get_code(code_sha256="…") (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 from a repository marked official are listed first. 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 four 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; or the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar). 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.

format_time brain-bzh/MCNN/defferard/utils.py community (archive-listed) ran · our draft was wrong fingerprinted MIT (permissive) · 8d7ec010c29e813b · report
lr_schedule Jeongyun-Lee-0423/Classification/cifar10_resnet.py community (archive-listed) ran · honoured contract fingerprinted no licence file found · pointer only · 67e18e3fdd805d03 · report
merge JoegameZhou/ms-resnetv2/model_utils/config.py community (archive-listed) ran · our draft was wrong Apache-2.0 (permissive) · b9e24c3e3e0e88a7 · report
ChebPreActResNet18 brain-bzh/MCNN/defferard/models/chebnet_preact_resnet.py community (archive-listed) unverified MIT (permissive) · cde0a8a6285fbdc7 · report
ChebPreActResNet34 brain-bzh/MCNN/defferard/models/chebnet_preact_resnet.py community (archive-listed) unverified MIT (permissive) · 7e7353108b3eed6f · report
GraphPreActResNet16 brain-bzh/MCNN/defferard/graph.py community (archive-listed) unverified MIT (permissive) · e653c1088e9d075a · report
GraphPreActResNet18 brain-bzh/MCNN/defferard/graph.py community (archive-listed) unverified MIT (permissive) · 1d4797c32ffbd2c7 · report
MiniChebNet brain-bzh/MCNN/defferard/models/chebnet_preact_resnet.py community (archive-listed) unverified MIT (permissive) · 9ade73802f8550c8 · report
batch_normalization_layer wenxinxu/resnet-in-tensorflow/resnet.py community (archive-listed) unverified MIT (permissive) · 6b5c23ab3faf0b82 · report
collect_images_from_directory sitegui/ceci-nest-pas-un-chat/myutils.py community (archive-listed) unverified MIT (permissive) · eff1e476cb4ffcbb · report
cov brain-bzh/MCNN/graph_processing/generate_graph.py community (archive-listed) unverified MIT (permissive) · b360fb65c97f182b · report
cre_groundtruth_dict JoegameZhou/ms-resnetv2/infer/resnetv2_101/sdk/classification_task_metric.py community (archive-listed) unverified Apache-2.0 (permissive) · 590865e4e0a3e889 · report
cre_groundtruth_dict_fromtxt JoegameZhou/ms-resnetv2/infer/resnetv2_101/sdk/classification_task_metric.py community (archive-listed) unverified Apache-2.0 (permissive) · b4b1e720f12820d2 · report
create_adjacence_matrix_from_covariance brain-bzh/MCNN/graph_processing/generate_graph.py community (archive-listed) unverified MIT (permissive) · 0ed3a7372ac14d78 · report
extract_data JoegameZhou/ms-resnetv2/infer/resnetv2_152/dataprocess/preprocess.py community (archive-listed) unverified Apache-2.0 (permissive) · 569e04900031b215 · report
extract_test_data JoegameZhou/ms-resnetv2/infer/resnetv2_152/dataprocess/preprocess.py community (archive-listed) unverified Apache-2.0 (permissive) · 847447d82fbd6630 · report
gen_file_name JoegameZhou/ms-resnetv2/infer/resnetv2_101/sdk/classification_task_metric.py community (archive-listed) unverified Apache-2.0 (permissive) · 0015055ff7835705 · report
generate_grid_graph brain-bzh/MCNN/graph_processing/translation_search.py community (archive-listed) unverified MIT (permissive) · 0430562f92bd5967 · report
get_center brain-bzh/MCNN/graph_processing/translation_search.py community (archive-listed) unverified MIT (permissive) · 43278cc51d5e4611 · report
get_mean_and_std brain-bzh/MCNN/defferard/utils.py community (archive-listed) unverified MIT (permissive) · 7bd3a77007a65eec · report
parse_cli_to_yaml JoegameZhou/ms-resnetv2/model_utils/config.py community (archive-listed) unverified Apache-2.0 (permissive) · 738456e1b7388e7e · report
parse_yaml JoegameZhou/ms-resnetv2/model_utils/config.py community (archive-listed) unverified Apache-2.0 (permissive) · 96a9939953d5992d · report
prepare_covariance_matrix brain-bzh/MCNN/graph_processing/generate_graph.py community (archive-listed) unverified MIT (permissive) · cfa5d19782f91d21 · report
read_file_list JoegameZhou/ms-resnetv2/infer/resnetv2_152/sdk/prec/eval_sdk.py community (archive-listed) unverified Apache-2.0 (permissive) · b5abc534653271c4 · report
read_graph brain-bzh/MCNN/graph_processing/translation_search.py community (archive-listed) unverified MIT (permissive) · 043a08fc8f5dbadb · report

Tasks

Image Classification

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Image Classification CIFAR-10 ResNet-1001 Percentage correct 95.4 #135 of 265 Archive leaderboard report
Image Classification CIFAR-100 ResNet-1001 Percentage correct 77.3 #143 of 211 Archive leaderboard report
Image Classification ImageNet ResNet-200 Top 1 Accuracy 79.9% #728 of 1060 Archive leaderboard report
Image Classification Kuzushiji-MNIST PreActResNet-18 Accuracy 97.82 #20 of 26 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 ConvolutionAffine CouplingAverage PoolingBatch NormalizationBottleneck Residual BlockConvolutionGlobal Average PoolingKaiming InitializationMax PoolingNormalizing FlowsRandom Horizontal FlipRandom Resized CropReLUResidual BlockResidual ConnectionSGD with MomentumStep DecayWeight Decay

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