AI Paper Atlas

Research map / Architectures and efficiency / Pruning and efficient tuning

Sparsity in Deep Learning: Pruning and growth for efficient inference and training in neural networks

Torsten Hoefler, Dan Alistarh, Tal Ben-Nun, Nikoli Dryden, Alexandra Peste

JMLR 2021 · Architectures and efficiency › Pruning and efficient tuning · approach: Sparse training from scratch · 536 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
replace magnitude pruning with second-order WoodFisher pruningResNet-50 gradual pruning to 95% sparsity · ImageNet valTop-1 accuracy: by 0.4% Top-1 validation accuracyImproved
add SGD fine-tuning between pruning stepsWoodFisher vs magnitude pruning · ImageNet val (ResNet-50)accuracy gap between methodsNo clear effect
replace SNIP with GraSP (gradient-flow-aware pruning)pruning at initialization · very sparse regimesaccuracyImproved
use SNIP at sparsity above 99%pruning at initialization · high sparsityaccuracyHurt
use iterative SNIP-it variant with more data in structure searchpruning before training · comparison with GraSPperformance cost: 5x improvement of performance over GraSPImproved

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