AI Paper Atlas

Research map / Architectures and efficiency / RNNs and normalization

SCAN: A Scalable Neural Networks Framework Towards Compact and Efficient Models

Linfeng Zhang, Zhanhong Tan, Jiebo Song, Jingwei Chen, Chenglong Bao, Kaisheng Ma

NeurIPS 2019 · Architectures and efficiency › RNNs and normalization · approach: Efficient network compression and training · 14 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 SCAN classifiers and ensemble their predictionsVGG16(BN) · CIFAR100accuracy (%): 72.46 -> 76.80Improved
Add SCAN classifiers and ensemble their predictionsVGG19(BN) · CIFAR100accuracy (%): 72.25 -> 75.43Improved
Add SCAN classifiers and ensemble their predictionsResNet18 · CIFAR100accuracy (%): 77.09 -> 80.46Improved
Add SCAN classifiers and ensemble their predictionsResNet50 · CIFAR100accuracy (%): 77.68 -> 81.78Improved
Add SCAN classifiers and ensemble their predictionsResNet101 · CIFAR100accuracy (%): 77.98 -> 82.06Improved
Add SCAN classifiers and ensemble their predictionsResNet152 · CIFAR100accuracy (%): 79.21 -> 82.94Improved
Add SCAN classifiers and ensemble their predictionsWRN20-8 · CIFAR100accuracy (%): 74.61 -> 80.04Improved
Add SCAN classifiers and ensemble their predictionsWRN44-8 · CIFAR100accuracy (%): 76.22 -> 80.35Improved
Add SCAN shallow classifiers and distillation trainingResNet18 · ImageNet (ILSVRC2012) validationaccuracy (%): 68.02 -> 69.32Improved
Add SCAN shallow classifiers and distillation trainingResNet50 · ImageNet (ILSVRC2012) validationaccuracy (%): 74.47 -> 75.88Improved
Add SCAN shallow classifiers and distillation trainingResNet101 · ImageNet (ILSVRC2012) validationaccuracy (%): 75.24 -> 76.32Improved
Add attention modules to shallow classifiers (ablation)ResNet50 with self distillation · CIFAR100accuracy (%): +5.46%, 4.13% ,and 5.16%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