AI Paper Atlas

Research map / Detection and segmentation / CNN features

Auto Learning Attention

Benteng Ma, Jing Zhang, Yong Xia, Dacheng Tao

NeurIPS 2020 · Detection and segmentation › CNN features · approach: Efficient network training and design · 17 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 searched HOGA attention (AutoLA)ResNet20 · CIFAR10accuracy: 91.95 -> 93.38Improved
add AutoLA instead of CBAMResNet50 · ImageNettop-1 error: 22.66 -> 21.82Improved
add AutoLAResNet50 SSD · COCOAP: 25.01 -> 27.78Improved
add AutoLA instead of SEResNet50 · COCO keypointsAP: 73.6 -> 74.6Improved
reduce group number from K=4 to K=2 (ablation)AutoLA · CIFAR10 ResNet20accuracy: 93.38 -> 93.18Hurt
use differentiable search instead of random search (ablation)HOGA · CIFAR10 ResNet32accuracy: 94.15 -> 94.33Improved
search HOGA directly on target backbone instead of transferring from ResNet56 (ablation)AutoLA · CIFAR10 ResNet20accuracy: 93.31 -> 93.38Improved
use searched HOGA instead of widened SE (ablation)ResNet20 · CIFAR10accuracy: 93.16 -> 93.38Improved
use searched ops instead of CBAM in all HOGA edges (ablation)HOGA k=4 · CIFAR10 ResNet20accuracy: 93.07 -> 93.38Improved

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