AI Paper Atlas

Research map / Architectures and efficiency / Pruning and efficient tuning

An Image Enhancing Pattern-based Sparsity for Real-time Inference on Mobile Devices

Xiaolong Ma, Wei Niu, Tianyun Zhang, Sijia Liu, Sheng Lin, Hongjia Li, Wujie Wen, Xiang Chen, Jian Tang, Kaisheng Ma, Bin Ren, Yanzhi Wang

ECCV 2020 · Architectures and efficiency › Pruning and efficient tuning · approach: Sparse training from scratch · 34 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
Prune VGG-16 to Phase 2 pattern sparsity at 8.0xVGG-16 · CIFAR-10accuracy (%): 93.5 -> 93.4Hurt
Apply pattern plus connectivity pruning at 8.0x to ResNet-18ResNet-18 · CIFAR-10accuracy (%): 94.0 -> 94.7Improved
Apply pattern plus connectivity pruning at 8.0x to ResNet-50ResNet-50 · CIFAR-10accuracy (%): 94.2 -> 95.2Improved
Raise ResNet-18 compression from 4.0x to 8.0xResNet-18 pattern pruning · ImageNettop-1 accuracy (%): 69.6 -> 67.1Hurt
Raise VGG-16 compression from 8.0x to 12.0xVGG-16 pattern pruning · ImageNettop-1 accuracy (%): 74.4 -> 73.6Hurt
Raise ResNet-50 compression from 3.9x to 5.8xResNet-50 pattern pruning · ImageNettop-1 accuracy (%): 75.9 -> 75.6Hurt
Raise VGG-16 compression from 8.0x to 19.7x (Phase 1)VGG-16 pattern pruning · CIFAR-10accuracy (%): 93.4 -> 93.2Hurt
Use the designed 8-pattern library (Phase 2) with connectivity sparsity instead of other libraries (ablation)pattern + connectivity pruning · ImageNet (ResNet-18, 6x)accuracyImproved
Use designed 8 patterns (Phase 2) on each convolution kernel (ablation)pattern pruning · CIFAR-10 and ImageNetaccuracyImproved
Pattern-prune VGG-16 with Phase 2 libraryVGG-16 · ImageNetvisual qualityImproved

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