AI Paper Atlas

Research map / Architectures and efficiency / CNN design

Channel Gating Neural Networks

Weizhe Hua, Yuan Zhou, Christopher M De Sa, Zhiru Zhang, G. Edward Suh

NeurIPS 2019 · Architectures and efficiency › CNN design · approach: Invertible and wide-network analysis · 133 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 dynamic channel gatingCNN architectures · CIFAR-10FLOP reduction: 2.7 - 8.0×Improved
Add channel grouping (ablation)CGNet-A · ImageNettop-1 accuracy drop: 1.3 -> 0.4Improved
Add channel grouping (ablation)CGNet-A · ImageNetFLOP reduction: $1.61\times$ -> $1.93\times$Improved
Add knowledge distillation (ablation)CGNet-A with channel grouping · ImageNettop-1 accuracy drop: 0.4 -> -0.5Improved
Add knowledge distillation (ablation)CGNet-A with channel grouping · ImageNetFLOP reduction: $1.93\times$ -> $2.55\times$Improved
Add knowledge distillation (ablation)CGNet · CIFAR-10accuracyNo clear effect
Use activation-wise and channel-wise gatesCGNet · CIFAR-10weight accesses: 3.1× and 4.4×Improved
Increase the target threshold TCGNet with channel-wise gate · CIFAR-10weight accesses: T equals 1.5 and 2.0Improved
Use channel grouping instead of a fixed base-path channel selection (ablation)CGNet · CIFAR-10top-1 accuracy: 0.8%Improved

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