AI Paper Atlas

Research map / Architectures and efficiency / CNN design

URNet: User-Resizable Residual Networks with Conditional Gating Module

Sangho Lee, Simyung Chang, Nojun Kwak

AAAI 2020 · Architectures and efficiency › CNN design · approach: Efficient CNN scaling and dynamic inference · 8 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 input-conditioned residual-block gates trained with scale lossResNet-110 · CIFAR-10, scale 0.6top-1 accuracy (%): 93.2 -> 93.7Improved
Use input-conditioned residual-block gates trained with scale lossResNet-101 · ImageNet validation, scale 0.8top-1 accuracy (%): 76.4 -> 76.4No clear effect
Use input-conditioned residual-block gates trained with scale lossFaster-RCNN (ResNet-101) · Pascal VOC 2007mAP: 74.96 -> 75.40Improved
Add binary-gate training alongside sigmoid-gate training (ablation)URNet-110 · CIFAR-10, scale 0.8top-1 accuracy (%): 81.7 -> 92.9Improved
Add sigmoid-gate training to binary-gate training (ablation)URNet-110 · CIFAR-10, scale 0.8resizabilityImproved
Replace per-block conditional gates with a separate ResNet-8 gate predictor (ablation)URNet-110 · CIFAR-10scale matchingHurt
Replace hard conditional gates with continuous block attention (ablation)URNet-110 · CIFAR-10, scale 0.8resizabilityHurt
Add previous-block features as CGM inputs (ablation)URNet-110 · CIFAR-10gate-usage variance: 0.00 -> 1.03Improved
Remove previous-block features from CGM inputs (ablation)URNet-110 · CIFAR-10, scale 0.8top-1 accuracy (%): 93.7 -> 93.5Hurt
Increase the scale-loss weight β too much (ablation)URNet-110 · CIFAR-10top-1 accuracyHurt
Train at a fixed target scale with scale annealing (ablation)URNet-101 · ImageNettop-1 accuracyImproved

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