AI Paper Atlas

Research map / Architectures and efficiency / CNN design

Learning to Learn Parameterized Classification Networks for Scalable Input Images

Duo Li, Anbang Yao, Qifeng Chen

ECCV 2020 · Architectures and efficiency › CNN design · approach: Filter design and trained-filter analysis · 11 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 scale distillation from large to small resolutions (ablation)SAN (MobileNetV2) · ImageNet val, train=test res 96top-1 accuracy: 61.86 -> 65.07Improved
use privatized BN layers per resolution instead of shared BN (ablation)SAN w/o SD (MobileNetV2) · ImageNet val, train=test res 96top-1 accuracy: 1.38 -> 61.86Improved
generate convolution weights with a meta learner instead of privatized convolution per resolution (ablation)SAN (ResNet-18) · ImageNet val, train=test res 96top-1 accuracy: 64.53 -> 65.30Improved
use a two-layer non-linear meta learner (8 hidden units) instead of linear (ablation)SAN (ResNet-18) · ImageNet val, train 224 test 64top-1 accuracy: 1.50Hurt
use a two-layer non-linear meta learner (8 hidden units) instead of linear (ablation)SAN (ResNet-18) · ImageNet val, medium resolutionstop-1 accuracy: 0.01No clear effect
double hidden units of non-linear meta learner to 16 (ablation)SAN (ResNet-18) · ImageNet val, train 192 test 64top-1 accuracy: 2.74Hurt
apply BN calibration for ideal inference at interpolated resolutions (ablation)SAN (ResNet-18) · ImageNet val, interpolated resolutionstop-1 accuracyImproved
use proxy inference instead of ideal inference with calibration (ablation)SAN (ResNet-18) · ImageNet val, test res 208/176/144/112top-1 accuracyNo clear effect
replace independently trained model with SANResNet-50 · ImageNet val, train 224 test 96top-1 accuracy: 70.61 -> 72.83Improved
replace independently trained model with SANResNet-18 · ImageNet val, train 224 test 64top-1 accuracy: 20.40 -> 34.98Improved
replace FixRes with SAN at large test resolutionResNet-50 (train 224) · ImageNet val, test res 384top-1 accuracy: 79.0 -> 77.0Hurt
replace FixRes with SAN at small test resolutionResNet-50 (train 224) · ImageNet val, test res 128top-1 accuracy: 67.7 -> 72.5Improved

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