AI Paper Atlas

Research map / Architectures and efficiency / Neural architecture search

MetaPruning: Meta Learning for Automatic Neural Network Channel Pruning

Zechun Liu, Haoyuan Mu, Xiangyu Zhang, Zichao Guo, Xin Yang, Kwang-Ting Cheng, Jian Sun

ICCV 2019 · Architectures and efficiency › Neural architecture search · approach: Latency-aware architecture search · 602 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 weight prediction (two FC layers) to the PruningNet (ablation)MetaPruning MobileNetV1 · ImageNet, under 45M FLOPstop-1 accuracy (%): 55.3 -> 1.9% higherImproved
add weight prediction to the PruningNet (ablation)PruningNet · ImageNet uniformly pruned MobileNetV1PruningNet accuracy: 10% lower accuracy withoutImproved
use MetaPruning instead of uniform pruningMobileNetV1 0.25x · ImageNet, FLOPs constrainttop-1 accuracy (%): 50.6 -> 57.2Improved
use MetaPruning instead of uniform pruningMobileNetV2 at 43M FLOPs · ImageNet, FLOPs constrainttop-1 accuracy (%): 54.6 -> 58.3Improved
use MetaPruning instead of uniform pruningResNet-50 at about 1G FLOPs · ImageNet, FLOPs constrainttop-1 accuracy (%): 72.0 -> 73.4Improved
use MetaPruning instead of uniform pruning under latencyMobileNetV1 0.25x · ImageNet, latency constrainttop-1 accuracy (%): 50.6 -> 59.6Improved
use MetaPruning instead of uniform pruning under latencyMobileNetV2 0.35x · ImageNet, latency constrainttop-1 accuracy (%): 54.6 -> 64.5Improved
use MetaPruning instead of AMCMobileNetV1 ~285M FLOPs · ImageNettop-1 accuracy (%): 70.5 -> 70.6Improved
use MetaPruning instead of AMCMobileNetV2 220M FLOPs · ImageNettop-1 accuracy (%): 70.8 -> 71.2Improved
use MetaPruning instead of ThiNetResNet-50 ~2G FLOPs · ImageNettop-1 accuracy (%): 74.7 -> 75.4Improved
prune shortcut channels jointly with middle channelsMobileNetV2 · ImageNet, FLOPs constrainttop-1 accuracy (%): 54.6 -> 58.3Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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