AI Paper Atlas

Research map / Architectures and efficiency / CNN design

Dynamic Dual Gating Neural Networks

Fanrong Li, Gang Li, Xiangyu He, Jian Cheng

ICCV 2021 · Architectures and efficiency › CNN design · approach: Efficient CNN scaling and dynamic inference · 58 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
use both spatial and channel gating instead of only one (ablation)DGNet · CIFAR-10 (ResNet-20/32)top-1 accuracyImproved
use spatial gating only instead of channel gating only (ablation)DGNet · CIFAR-10 (ResNet-20/32)top-1 accuracyMixed
use 1x1 fine-grained spatial tiles instead of 2x2 or hybrid tiles (ablation)DGNet · CIFAR-10 (ResNet-20/32)top-1 accuracyHurt
use 8x8 tiles in all layers instead of hybrid tile sizes (ablation)DGNet · CIFAR-10 (ResNet-20/32)top-1 accuracyHurt
use hybrid tile sizes (per-stage) instead of uniform tile size (ablation)DGNet · CIFAR-10 (ResNet-20/32)top-1 accuracyImproved
avoid applying spatial and channel gating in the same layer (proposed integration) instead of inserting spatial gating in all convolutions (ablation)DGNet · CIFAR-10 (ResNet-20/32)top-1 accuracyImproved
use a shared tiled mask (t=4) instead of a fine-grained mask (t=1) when dilating for 3x3 convolutions (ablation)spatial gating mask · 28x28 mask visualizationmask sparsity after dilation: 47.84% -> 0.26% for t=1; 44.90% -> 23.85% for t=4Improved
apply dual gating with 50% target to ResNet-18ResNet-18 · ImageNet valtop-1 accuracy: 69.76 -> 70.12Improved
apply dual gating with 60% target to ResNet-50ResNet-50 · ImageNet valtop-1 accuracy: 76.13 -> 76.41Improved
raise target sparsity from 60% to 70%DGNet ResNet-34 · ImageNet valtop-1 accuracy: 73.01 -> 71.95Hurt
apply dual gating (60%) to RetinaNet backboneRetinaNet · COCO 2017AP0.5:0.95: 36.5 -> 36.1No clear effect
apply dual gating (60%) to Faster-RCNN backboneFaster-RCNN FPN · COCO 2017AP0.5:0.95: 37.4 -> 37.2No clear effect

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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