AI Paper Atlas

Research map / Architectures and efficiency / Pruning and efficient tuning

Channel Pruning for Accelerating Very Deep Neural Networks

Yihui He, Xiangyu Zhang, Jian Sun

ICCV 2017 · Architectures and efficiency › Pruning and efficient tuning · approach: Structured network pruning · 2,753 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 fine-tuning after channel pruning (ablation)channel pruning · ImageNet VGG-16 4xincrease of top-5 error (%): 7.9 -> 1.0Improved
add fine-tuning after channel pruning (ablation)channel pruning · ImageNet VGG-16 5xincrease of top-5 error (%): 22.0 -> 1.7Improved
add fine-tuning after channel pruning (ablation)channel pruning · ImageNet VGG-16 2xincrease of top-5 error (%): 2.7 -> 0Improved
add multi-branch enhancement and fine-tuning (ablation)channel pruning · ImageNet ResNet-50 2xincrease of top-5 error (%): 8.0 -> 4.0 -> 1.4Improved
prune with reconstruction-based channel selection instead of first k or max response (ablation)channel pruning · VGG-16 single layerincrease of top-5 errorImproved
select channels by max response (filter weight magnitude) instead of first k (ablation)channel pruning · VGG-16 single layerincrease of top-5 errorNo clear effect
combine channel pruning with spatial and channel factorization (3C)VGG-16 · ImageNet 4xincrease of top-5 error (%): 1.0 -> 0.0Improved
prune channels instead of train same architecture from scratchVGG-16 4x · ImageNetincrease of top-5 error (%): 1.8 -> 1.0Improved
prune channels instead of uniformly halving filters and training from scratchVGG-16 4x · ImageNetincrease of top-5 error (%): 2.4 -> 1.0Improved
channel pruning vs Filter pruning (Li et al.)VGG-16 · ImageNet 5xincrease of top-5 error (%): 14.6 -> 1.7Improved
channel pruning on Faster R-CNN backbone 2xVGG-16 · VOC07mAP: 68.7 -> 68.3No clear effect
channel pruning on Faster R-CNN backbone 4xVGG-16 · VOC07mAP: 68.7 -> 66.9Hurt

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