Papers › Compounding the Performance Improvements of Assembled Techniques in a Convolutional...

Compounding the Performance Improvements of Assembled Techniques in a Convolutional Neural Network

17 Jan 2020arXiv:2001.06268archive 2025-07-28

Jungkyu Lee, Taeryun Won, Tae Kwan Lee, Hyemin Lee, Geonmo Gu, Kiho Hong

Recent studies in image classification have demonstrated a variety of techniques for improving the performance of Convolutional Neural Networks (CNNs). However, attempts to combine existing techniques to create a practical model are still uncommon. In this study, we carry out extensive experiments to validate that carefully assembling these techniques and applying them to basic CNN models (e.g. ResNet and MobileNet) can improve the accuracy and robustness of the models while minimizing the loss of throughput. Our proposed assembled ResNet-50 shows improvements in top-1 accuracy from 76.3% to 82.78%, mCE from 76.0% to 48.9% and mFR from 57.7% to 32.3% on ILSVRC2012 validation set. With these improvements, inference throughput only decreases from 536 to 312. To verify the performance improvement in transfer learning, fine grained classification and image retrieval tasks were tested on several public datasets and showed that the improvement to backbone network performance boosted transfer learning performance significantly. Our approach achieved 1st place in the iFood Competition Fine-Grained Visual Recognition at CVPR 2019, and the source code and trained models are available at https://github.com/clovaai/assembled-cnn

PaperPDFCodeCode Syntology ran

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

Syntology 11 code samples harvested from 1 repository linked to this paper: 3 ran (of which 0 constructed an object rather than computing a result; 2 with no instrument failure: 0 honoured, 0 violated, 2 with no contract checked; 1 where Syntology's instrument failed) · 8 unverified. Where Syntology's instrument failed: 1 where the contract check could not run; failures of Syntology's instrument, not of the code.

By repository: official repository: 11 samples from 1 repository, 3 ran (2 with no instrument failure; 1 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.

Levels written between 2026-09-22 and 2026-09-25 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed.

clovaai/assembled-cnn officialmentioned in papermentioned on GitHubtfApache-2.0Syntology: 11 samples, 3 ran (1 instrument-failed) · 8 unverified 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

11 samples harvested: 3 ran (of which 0 constructed an object rather than computing a result; 2 with no instrument failure: 0 honoured, 0 violated, 2 with no contract checked; 1 where Syntology's instrument failed) · 8 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Syntology's record for this page has not changed since , the first build that kept a record date for it; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

1ran; the contract check could not run
2ran
8unverified

Licence: 0 of the 11 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 clovaai/assembled-cnn. “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.

clovaai/assembled-cnn official repository 11 samples, 3 ran (1 instrument-failed) · 8 unverified
fixed_padding in 28 papers nets/model_helper.py ran; the contract check could not run (RAISES) · metamorphic tier: invariant fingerprinted Apache-2.0 (permissive) · 5756d6bfbcce2dfc · report fixture: signature inputs:EagerTensor/4/float, kernel_size:int/0/int, data_format:str/0/str · returned EagerTensor · level recorded 2026-09-22
get_block_sizes functions/model_fns.py ran Apache-2.0 (permissive) · 8ca9aa61ab881fc5 · report fixture: signature resnet_size:int/0/int, resnet_version:int/0/int · returned list · level recorded 2026-09-25
keep_prob_decay functions/model_fns.py ran Apache-2.0 (permissive) · 3a511a9f3d42665c · report fixture: signature starter_kp:float/0/float, end_kp:float/0/float, decay_steps:int/0/int · returned function · level recorded 2026-09-25
anti_aliased_downsample nets/blocks.py unverified Apache-2.0 (permissive) · 0970ec81afd374c8 · report
batch_norm in 46 papers nets/model_helper.py unverified Apache-2.0 (permissive) · aa9fde2ef364ebff · report
conv2d_fixed_padding in 12 papers nets/model_helper.py unverified Apache-2.0 (permissive) · c756a6986ed2d1b8 · report
generalized_mean_pooling nets/blocks.py unverified Apache-2.0 (permissive) · bc3aa92a65ffc867 · report
get_sup_loss losses/cls_losses.py unverified Apache-2.0 (permissive) · 58ddea37201045ea · report
get_train_op nets/optimizer_setting.py unverified Apache-2.0 (permissive) · 8e9e9366b2d2231c · report
learning_rate_with_decay functions/model_fns.py unverified Apache-2.0 (permissive) · d9ed2858fbdb4769 · report
sk_conv2d nets/blocks.py unverified Apache-2.0 (permissive) · 58b86e2110d9f973 · report

Tasks

Fine-Grained Image ClassificationFine-Grained Visual RecognitionGeneral ClassificationImage ClassificationImage RetrievalRetrievalTransfer Learningimage-classification

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Fine-Grained Image Classification FGVC Aircraft Assemble-ResNet-FGVC-50 Accuracy 92.4 #41 of 57 Archive leaderboard report
Fine-Grained Image Classification Food-101 Assemble-ResNet-FGVC-50 Accuracy 92.5 #9 of 15 Archive leaderboard report
Fine-Grained Image Classification Food-101 Assemble-ResNet-FGVC-50 Top 1 Accuracy 92.47 #9 of 15 Archive leaderboard report
Fine-Grained Image Classification Oxford 102 Flowers Assemble-ResNet Accuracy 98.9% #9 of 25 Archive leaderboard report
Fine-Grained Image Classification Oxford-IIIT Pets Assemble-ResNet-FGVC-50 Accuracy 94.3% #6 of 19 Archive leaderboard report
Fine-Grained Image Classification Oxford-IIIT Pets Assemble-ResNet-FGVC-50 Top-1 Error Rate 5.7 #6 of 19 Archive leaderboard report
Fine-Grained Image Classification SOP Assemble-ResNet-FGVC-50 Recall@1 85.9 #1 of 1 Archive leaderboard report
Fine-Grained Image Classification Stanford Cars Assemble-ResNet-FGVC-50 Accuracy 94.4% #47 of 83 Archive leaderboard report
Image Classification ImageNet Assemble-ResNet152 GFLOPs 15.8 #332 of 1060 Archive leaderboard report
Image Classification ImageNet Assemble-ResNet152 Top 1 Accuracy 84.2% #332 of 1060 Archive leaderboard report
Image Classification ImageNet ReaL Assemble-ResNet152 Accuracy 88.65% #27 of 57 Archive leaderboard report
Image Classification ImageNet ReaL Assemble ResNet-50 Accuracy 87.82% #31 of 57 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 ConvolutionAnti-Alias DownsamplingAssemble-ResNetAutoAugmentAverage PoolingBatch NormalizationBig-Little ModuleBottleneck Residual BlockColorJitterConvolutionCosine AnnealingCutoutDense ConnectionsDepthwise ConvolutionDepthwise Separable ConvolutionDilated ConvolutionDropBlockGlobal Average PoolingKaiming InitializationLSTMLabel SmoothingLinear LayerLinear Warmup With Cosine AnnealingMax PoolingMixupMobileNetV1Pointwise ConvolutionRandom Horizontal FlipRandom Resized CropReLUResNet-DResidual BlockResidual ConnectionSGD with MomentumSelective KernelSelective Kernel ConvolutionSigmoid ActivationSoftmaxTanh ActivationWeight DecayXavier Initialization

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