AI Paper Atlas

Research map / Architectures and efficiency / Pruning and efficient tuning

Coordinating Filters for Faster Deep Neural Networks

Wei Wen, Cong Xu, Chunpeng Wu, Yandan Wang, Yiran Chen, Hai Li

ICCV 2017 · Architectures and efficiency › Pruning and efficient tuning · approach: Low-rank compression and merging · 128 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 l2-norm Force Regularization during trainingAlexNet PCA LRA · AlexNet (ImageNet)average rank ratio (%): 72.29% -> 46.98%Improved
add l1-norm Force Regularization during trainingAlexNet PCA LRA · AlexNet (ImageNet)average rank ratio (%): 72.29% -> 50.03%Improved
add l2-norm Force Regularization during trainingConvNet · ConvNet (CIFAR-10)average rank ratio (%): 74.48% -> 54.17%Improved
add l2-norm Force Regularization before PCA decompositionAlexNet lra · AlexNet conv3 (GPU)speedup: 1.58x -> 2.16xImproved
add l2-norm Force Regularization before PCA decompositionAlexNet lra · AlexNet conv4 (CPU)speedup: 1.60x -> 2.76xImproved
add l2-norm Force Regularization on conv3-5 of lra2AlexNet lra2 · AlexNet ImageNet GPU conv3speedup: 1.89x -> 2.25xImproved
use lra2 (3x3 shape preserved) instead of lra1 for conv3-5AlexNet · AlexNet ImageNet GPU conv3speedup: 0.86x -> 1.89xImproved
scale regularization gradient by ||W_i|| instead of 1/||W_i|| (ablation)Force Regularization · ConvNet (CIFAR-10)conv3 rank (of 64): 32/64 -> 30/64Improved
use PCA or SVD instead of k-means for decompositionAlexNet with l2-norm force · AlexNet (ImageNet)top-1 error (%): 44.80% -> 43.25%Improved
add Discrimination Regularization (reverse sign force) with l2-norm forceAlexNet · AlexNet (ImageNet)top-1 error (%): 42.63% -> 41.71%Improved
add Discrimination Regularization (reverse sign force) with l2-norm forceResNets-20 · CIFAR-10top-1 error (%): 8.82% -> 7.97%Improved
add Force Regularization before LRA when comparing at similar top-5 error to SSLAlexNet lra2 · AlexNet ImageNet CPUspeedup: 3.13x -> 4.05xImproved

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