AI Paper Atlas

Research map / Architectures and efficiency / CNN design

Convolutional Neural Networks with Gated Recurrent Connections

Jianfeng Wang, Xiaolin Hu

TPAMI 2021 · Architectures and efficiency › CNN design · approach: Compact CNN architectures · 66 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add gates on recurrent connections (RCNN -> original GRCNN), T=5, tied weights (ablation)RCNN · CIFAR-10test error (%): 5.88 -> 5.57Improved
add gates, T=2, tied weights (ablation)RCNN · CIFAR-10test error (%): 5.81 -> 5.92Hurt
add accumulated contributions over iterations (original -> improved GRCNN), T=5 untied (ablation)GRCNN · CIFAR-10test error (%): 5.23 -> 5.16Improved
increase iterations T from 2 to 5 in each GRCL, tied weights (ablation)original GRCNN · CIFAR-10test error (%): 5.92 -> 5.57Improved
increase iterations T from 2 to 5, tied weights (ablation)improved GRCNN · CIFAR-10test error (%): 5.83 -> 5.49Improved
increase iterations T from 2 to 5, tied weights (ablation)RCNN · CIFAR-10test error (%): 5.65 -> 5.88Hurt
untie weights across iterations, T=5 (ablation)original GRCNN · CIFAR-10test error (%): 5.57 -> 5.23Improved
add selective kernel convolution to GRCNN-110GRCNN-110 · CIFAR-10test error (%): 3.82 -> 3.42Improved
replace SK block in SKNet-110 with gated recurrent SK-GRCNN-110SKNet-110 · CIFAR-100test error (%): 18.34 -> 18.18Improved
replace ResNet-50 with GRCNN-55ResNet-50 · ImageNet valtop-1 error (%): 24.70 -> 23.12Improved
add deformable v2 convolution to GRCNN-55GRCNN-55 · ImageNet valtop-1 error (%): 23.12 -> 21.63Improved
add recurrent iterations T=1 to T=3 with gates (ablation)GRCNN · IIIT5Kaccuracy (%): 78.63 -> 81.01Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy