AI Paper Atlas

Research map / Architectures and efficiency / CNN design

MUXConv: Information Multiplexing in Convolutional Neural Networks

Zhichao Lu, Kalyanmoy Deb, Vishnu Naresh Boddeti

CVPR 2020 · Architectures and efficiency › CNN design · approach: Efficient CNN scaling and dynamic inference · 63 citations (OpenAlex)

Check 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 spatial multiplexing to depth-wise separable convolutions (ablation)MobileNetV2 · ImageNet at low-MAdds model sizesTop-1 accuracy: 5.8%Improved
Add spatial multiplexing to depth-wise separable convolution layers (ablation)MobileNetV2 · ImageNet at fixed spatial resolutionTop-1 accuracyImproved
Leave out 25% of input channels in normal blocks (ablation)MobileNetV2 with channel multiplexing · ImageNetParameters: 13%Improved
Leave out 25% of input channels in normal blocks (ablation)MobileNetV2 with channel multiplexing · ImageNetMAdds: 20%Improved
Leave out 25% of input channels in normal blocks (ablation)MobileNetV2 with channel multiplexing · ImageNetPredictive accuracyNo clear effect
Solve preference-decomposed subproblems collectively instead of one at a time (ablation)PBI-based architecture search · NASBench101Search effectiveness under fixed budgetImproved
Use MUXNet-m instead of MobileNetV2 as the SSDLite backboneSSDLite · PASCAL VOC 2007 testmAP: 67.4 -> 68.6Improved
Use MUXNet-m instead of MobileNetV2 as the PPM backbonePPM segmentation · ADE20K validationmIoU: 35.76 -> 35.80No clear effect
Use MUXNet-m instead of CheXNetChestX-Ray14 transfer learning · ChestX-Ray14 testAUROC: 84.4 -> 84.1No clear effect
Use MUXNet-m instead of MobileNetV2ADE20K segmentation with C1 head · ADE20K validationmIoU: 34.84 -> 32.42Hurt
Use MUXNet-m instead of MobileNetV2CIFAR transfer learning · CIFAR-10 testTop-1 accuracy: 95.74 -> 98.00Mixed

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