Papers › Inception-v4, Inception-ResNet and the Impact of Residual Connections on Learning

Inception-v4, Inception-ResNet and the Impact of Residual Connections on Learning

23 Feb 2016arXiv:1602.07261archive 2025-07-28

Christian Szegedy, Sergey Ioffe, Vincent Vanhoucke, Alex Alemi

Very deep convolutional networks have been central to the largest advances in image recognition performance in recent years. One example is the Inception architecture that has been shown to achieve very good performance at relatively low computational cost. Recently, the introduction of residual connections in conjunction with a more traditional architecture has yielded state-of-the-art performance in the 2015 ILSVRC challenge; its performance was similar to the latest generation Inception-v3 network. This raises the question of whether there are any benefit in combining the Inception architecture with residual connections. Here we give clear empirical evidence that training with residual connections accelerates the training of Inception networks significantly. There is also some evidence of residual Inception networks outperforming similarly expensive Inception networks without residual connections by a thin margin. We also present several new streamlined architectures for both residual and non-residual Inception networks. These variations improve the single-frame recognition performance on the ILSVRC 2012 classification task significantly. We further demonstrate how proper activation scaling stabilizes the training of very wide residual Inception networks. With an ensemble of three residual and one Inception-v4, we achieve 3.08 percent top-5 error on the test set of the ImageNet classification (CLS) challenge

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="1602.07261")

Code

Syntology Ran 1 of 1 code samples harvested from 1 repository linked to this paper; 0 have no recorded run. Of those that ran: 1 ran · our draft was wrong.

By repository: community (archive-listed): 1 sample from 1 repository, 1 ran. The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

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

0jason000/Inception_V4 mentioned on GitHubmindspore report
2023-MindSpore-1/ms-code-204 mentioned on GitHubmindspore report
32shivang/Blind-Eye mentioned on GitHubtf report
92coorob/facerec2 mentioned on GitHubtf report
Dhruvpatel2112/Dhruv-Patel mentioned on GitHubtf report
IMvision12/keras-vision-models mentioned on GitHubpytorch report
Liuyubao/transfer-learning mentioned on GitHubtf report
Mariya1285/Codeathon mentioned on GitHubtf report
Mind23-2/MindCode-54 mentioned on GitHubmindspore report
Wenzhima66/readme mentioned on GitHubpytorch report
baldassarreFe/deep-koalarization mentioned on GitHubtfMIT report
chldydgh4687/2020-1.VideoCaptioning mentioned on GitHubpytorch report
hujinxinb/face_detect mentioned on GitHubtf report
kentsommer/keras-inceptionV4 mentioned on GitHubtf report
khazit/Lip2Word mentioned on GitHubtf report
khazit/LipNet mentioned on GitHubtf report
kingcong/gpu_inceptionv4 mentioned on GitHubmindspore report
kobiso/CBAM-tensorflow mentioned on GitHubtf report
kobiso/CBAM-tensorflow-slim mentioned on GitHubtfMIT report
lyc760214/ML_DL mentioned on GitHub report
marload/ConvNets-TensorFlow2 mentioned on GitHubtf report
mehmetPazar/GraduationProject mentioned on GitHubtf report
nknytk/ml-study mentioned on GitHub report
poloclub/jpeg-defense mentioned on GitHubtfMIT report
rickyHong/JPEG-Defense-repl mentioned on GitHubtfMIT report
rubenrosales/inception_v4 mentioned on GitHub report
skaldek/InceptionResNetV2 mentioned on GitHubtf report
systemcorp-ai/InceptionV4 mentioned on GitHubtf report
syzroy/4099-Emotion-Analyser mentioned on GitHub report
tensorflow/models mentioned on GitHubtf report
tensorflow/models mentioned on GitHubtf report
titu1994/inception-v4 mentioned on GitHub report
vudung45/FaceRec mentioned on GitHubtf report
xlsarath/Face-Recognition mentioned on GitHubtf report
yang-neu/FaceRec mentioned on GitHubtf report
yangyucheng000/ascend_inceptionv4 mentioned on GitHubmindspore report
yangyucheng000/inceptionv4 mentioned on GitHubmindspore report
zjZSTU/GoogLeNet mentioned on GitHubpytorch report
PaddlePaddle/PaddleClas paddleApache-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

1 sample harvested; 1 ran; 0 honoured the contract we drafted; 0 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 · our draft was wrong

Licence: 0 of the 1 sample 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 kobiso/CBAM-tensorflow. “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.

parse_arguments kobiso/CBAM-tensorflow/Inception_v4.py community (archive-listed) ran · our draft was wrong MIT (permissive) · f891cdc1c30186df · report

Tasks

ClassificationGeneral ClassificationImage Classification

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Classification InDL Inception ResNet V2 Average Recall 90.27% #4 of 9 Archive leaderboard report
Image Classification ImageNet Inception ResNet V2 Number of params 55.8M #719 of 1060 Archive leaderboard report
Image Classification ImageNet Inception ResNet V2 Top 1 Accuracy 80.1% #719 of 1060 Archive leaderboard report
Image Classification OmniBenchmark InceptionV4 Average Top-1 Accuracy 32.3 #16 of 22 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

Introduced by this paper: Inception-A, Inception-B, Inception-C, Inception-ResNet-v2 Reduction-B, Inception-ResNet-v2-B, Inception-ResNet-v2-C, Inception-v4, Reduction-A, Reduction-B

1x1 ConvolutionAuxiliary ClassifierAverage PoolingBatch NormalizationBottleneck Residual BlockConvolutionDense ConnectionsDropoutExponential DecayGlobal Average PoolingInception-AInception-BInception-CInception-ResNet-v2Inception-ResNet-v2 Reduction-BInception-ResNet-v2-AInception-ResNet-v2-BInception-ResNet-v2-CInception-v3Inception-v3 ModuleInception-v4Kaiming InitializationLabel SmoothingMax PoolingRMSPropReLUReduction-AReduction-BResidual BlockResidual ConnectionSoftmax

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