AI Paper Atlas

Research map / Architectures and efficiency / Pruning and efficient tuning

Practical Network Acceleration With Tiny Sets: Hypothesis, Theory, and Algorithm

Guo-Hua Wang, Jianxin Wu

TPAMI 2024 · Architectures and efficiency › Pruning and efficient tuning · approach: Joint pruning and quantization · 6 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
drop blocks instead of pruning filters (BP baseline)ResNet-34 finetuned with cross-entropy · ImageNet-1k, 16% latency, 50 imagesTop-1 accuracy (%): 39.0 -> 66.5Improved
drop blocks instead of pruning filters (BP baseline)ResNet-34 finetuned with cross-entropy · ImageNet-1k, 22.1% latency, 50 imagesTop-1 accuracy (%): 24.2 -> 60.6Improved
use Practise (recoverability-based block selection plus feature mimicking) instead of MiRResNet-34 · ImageNet-1k, 16% latency, 50 imagesTop-1 accuracy (%): 64.1 -> 70.3Improved
use Practise instead of MiR at larger latency reductionResNet-34 · ImageNet-1k, 22.1% latency, 500 imagesTop-1 accuracy (%): +3.6%Improved
use Practise instead of MiR at small latency reductionResNet-34 · ImageNet-1k, 8.2% latency, 500 imagesTop-1 accuracy (%): +1.1%Improved
use Practise instead of MiR on MobileNetV2MobileNetV2 · ImageNet-1k, ~16% latency, 500 imagesTop-1 accuracy (%): 67.6 -> 69.3Improved
replace filter-pruned data-free baseline ADI* with Practise on synthesized imagesResNet-50 · ImageNet-1k, data-freeTop-1 accuracy (%): 73.5 -> 74.8Improved
replace filter-pruned data-free baseline ADI* with Practise on synthesized imagesMobileNetV2 · ImageNet-1k, data-freeTop-1 accuracy (%): 62.8 -> 68.0Improved
add recoverability as block-selection criterion (vs CURL) with 5 dropped blocks (ablation)Practise · ImageNet-1k, ResNet-34, 500 imagesTop-1 accuracy (%): 64.97 -> 69.35Improved
add recoverability as block-selection criterion (vs L2 distance) with 4 dropped blocks (ablation)Practise · ImageNet-1k, ResNet-34, 500 imagesTop-1 accuracy (%): 68.61 -> 70.86Improved
add recoverability criterion; gap to other criteria with 1 dropped block (ablation)Practise · ImageNet-1k, ResNet-34, 500 imagesTop-1 accuracy (%): 72.33 -> 73.02Improved
increase out-of-domain training images from 50 to all (Place365)Practise · ImageNet-1k, ResNet-50 pruned 21%Top-1 accuracy (%): 72.80 -> 75.21Improved

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