AI Paper Atlas

Research map / Architectures and efficiency / Pruning and efficient tuning

Exposing and Exploiting Fine-Grained Block Structures for Fast and Accurate Sparse Training

Peng Jiang, Lihan Hu, Shihui Song

NeurIPS 2022 · Architectures and efficiency › Pruning and efficient tuning · approach: Sparse training from scratch · 0 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
remove model adaptation (static shuffled block, SSB-16) at 0.5 sparsity (ablation)DSB-16 · ImageNet ResNet50top-1 accuracy: 75.83 -> 76.33Improved
remove model adaptation (SSB-16) at 0.5 sparsity (ablation)DSB-16 · ImageNet ResNet18top-1 accuracy: 68.52 -> 69.84Improved
remove model adaptation (SSB-16) at 0.75 sparsity (ablation)DSB-16 · ImageNet ResNet50top-1 accuracy: 72.76 -> 74.04Improved
remove model adaptation (SSB-16) at 0.75 sparsity (ablation)DSB-16 · ImageNet ResNet18top-1 accuracy: 64.49 -> 65.72Improved
DSB-16 at 0.5 sparsity versus dense trainingDSB-16 · ImageNet ResNet50top-1 accuracy: 76.13 -> 76.33Improved
DSB-16 at 0.75 sparsity versus SSLT at 0.7 sparsityDSB-16 · ImageNet ResNet50top-1 accuracy: 71.5 -> 74.04Improved
DSB-16 at 0.5 sparsity versus SSLT at 0.51 sparsityDSB-16 · ImageNet ResNet50top-1 accuracy: 75.65 -> 76.33Improved
use block structure (DSB) instead of unstructured RigL (block size 1)DSB · CIFAR10test accuracyHurt
remove shuffling (nonshuffled block sparsity, DNSB) (ablation)DSB · CIFAR10/100test accuracyImproved
remove model adaptation (SSB) on CIFAR10/100 (ablation)DSB · CIFAR10/100test accuracyImproved
block size 4 or 8 or 16 at sparsity up to 0.8 (ablation)DSB · CIFAR10test accuracyNo clear effect
increase block size to 32 (ablation)DSB · CIFAR10test accuracyHurt

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