AI Paper Atlas

Research map / Architectures and efficiency / RNNs and normalization

Dynamic ReLU

Yinpeng Chen, Xiyang Dai, Mengchen Liu, Dongdong Chen, Lu Yuan, Zicheng Liu

ECCV 2020 · Architectures and efficiency › RNNs and normalization · approach: Width and depth of ReLU networks · 194 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
replace static ReLU with DY-ReLU-BMobileNetV2 x1.0 · ImageNet valtop-1 accuracy: 72.0 -> 76.2Improved
replace SE and h-swish with DY-ReLUMobileNetV3-Small · ImageNet valtop-1 accuracy: 67.4 -> 69.7Improved
replace ReLU with DY-ReLUResNet-10 · ImageNet valtop-1 accuracy: 63.0 -> 66.3Improved
replace ReLU with DY-ReLU-C in backbone and headMobileNetV2 x1.0 · COCO keypoint val2017AP: 64.6 -> 68.1Improved
use channel-wise DY-ReLU-B instead of channel-shared DY-ReLU-A (ablation)DY-ReLU · ImageNet val (MobileNetV2 x0.35)top-1 accuracy: 63.28 -> 66.36Improved
add spatial-wise activation (DY-ReLU-C instead of B) for classification (ablation)DY-ReLU · ImageNet val (MobileNetV2 x0.35)top-1 accuracy: 66.36 -> 66.31No clear effect
use DY-ReLU-C instead of DY-ReLU-A in backbone (ablation)MobileNetV2 x0.5 keypoint · COCO val2017AP: 58.97 -> 62.23Improved
use DY-ReLU-C instead of DY-ReLU-B in the head (ReLU backbone) (ablation)MobileNetV2 x0.5 keypoint · COCO val2017AP: 58.72 -> 61.03Improved
use DY-ReLU-C in both backbone and head (ablation)MobileNetV2 x0.5 keypoint · COCO val2017AP: 59.26 -> 63.27Improved
use DY-ReLU-A in the head (ablation)MobileNetV2 x0.5 keypoint · COCO val2017AP: 59.26 -> 57.12Hurt
make first segment dynamic (slope a_c(x)) (ablation)DY-ReLU · ImageNet val (MobileNetV2 x0.35)top-1 accuracy: 60.3 -> 63.8Improved
add a second dynamic segment (K=2 max) (ablation)DY-ReLU · ImageNet val (MobileNetV2 x0.35)top-1 accuracy: 63.8 -> 65.7Improved

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