AI Paper Atlas

Research map / Architectures and efficiency / Pruning and efficient tuning

Stable Low-rank Tensor Decomposition for Compression of Convolutional Neural Network

Anh-Huy Phan, Konstantin Sobolev, Konstantin Sozykin, Dmitry Ermilov, Julia Gusak, Petr Tichavský, Valeriy Glukhov, Ivan Oseledets, Andrzej Cichocki

ECCV 2020 · Architectures and efficiency › Pruning and efficient tuning · approach: Low-rank compression and merging · 150 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 error-preserving sensitivity correction (EPC) to CPD of kernels (ablation)CPD-compressed ResNet-18 · ILSVRC-12, ResNet-18 layerstop-1 accuracy: ~2% top-1 difference on some deep layersImproved
add EPC correction to CPD at rank 500 (ablation)CPD-compressed ResNet-18 layer4.1.conv1 · ILSVRC-12top-1 accuracy: rank 450 -> original accuracy; rank 250 -> less than 1% lossImproved
add EPC correction to CPD (ablation)CPD-compressed ResNet-18 · ILSVRC-12 and CIFAR-100sensitivity of decompositionImproved
replace CPD-EPC with TKD-CPD-EPC hybrid at small ranksResNet-18 layer 4.0.conv1 · CIFAR-100top-1 accuracyHurt
replace CPD-EPC with TKD-CPD-EPC hybrid at CP rank >= 110ResNet-18 layer 4.0.conv1 · CIFAR-100top-1/top-5 accuracy: R_CP >= 110Improved
replace CPD-EPC with TKD-CPD-EPCResNet-18 single layers · CIFAR-100accuracy vs FLOPs/parametersImproved
compress ResNet-50 whole network with oursResNet-50 · ILSVRC-12inference time on CPU: 3.92 -> 2.84Improved
compress ResNet-50 whole network with oursResNet-50 · ILSVRC-12inference time on V100 GPU: 102.3 -> 89.5Improved
compress ResNet-50 whole network with oursResNet-50 · ILSVRC-12inference time on Snapdragon 845: 221 -> 171Improved
our method vs MUSCO on ResNet-18ResNet-18 · ILSVRC-12FLOPs reduction and top-1 drop: 2.42 -> 3.09 FLOPs; -0.47 -> -0.69 top-1Mixed

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