AI Paper Atlas

Research map / Architectures and efficiency / Pruning and efficient tuning

Connectivity Matters: Neural Network Pruning Through the Lens of Effective Sparsity

Artem Vysogorets, Julia Kempe

JMLR 2023 · Architectures and efficiency › Pruning and efficient tuning · approach: Structured network pruning · 15 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
evaluate with effective instead of direct compression for pruning methodsSynFlow vs SNIP and GraSP · LeNet-300-100 / MNISTtest accuracy vs compressionMixed
evaluate with effective instead of direct compression on convolutional netsSynFlow vs other methods · VGG-19 and ResNet-18performance gapMixed
use LAMP pruningLAMP · all architectureseffective vs direct compressionHurt
use IGQ layerwise sparsity for random pruning instead of Uniform/Uniform+ (ablation)random pruning · ResNet-50 up to 99% sparsitytest accuracyImproved
use ERK instead of IGQ for random pruning (ablation)random pruning · VGG-19 at 1000x or higher compressiontest accuracyImproved
use IGQ layerwise quotas in magnitude pruning instead of Uniform/ERK baselines (ablation)magnitude pruning with retraining · all architectures, up to 100xtest accuracyImproved
use global magnitude pruning instead of IGQ (ablation)magnitude pruning · VGG-19test accuracyHurt
use effective pruning (binary search to target effective sparsity) (ablation)ranking-based and random pruning · ResNet-18reaching target effective sparsityImproved

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