AI Paper Atlas

Research map / Architectures and efficiency / Pruning and efficient tuning

Accelerating CNN Training by Pruning Activation Gradients

Xucheng Ye, Pengcheng Dai, Junyu Luo, Xin Guo, Yingjie Qi, Jianlei Yang, Yiran Chen

ECCV 2020 · Architectures and efficiency › Pruning and efficient tuning · approach: Structured network pruning · 40 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
raise pruning rate to p=99% (ablation)stochastic gradient pruning, AlexNet · ImageNettraining accuracy (%): 56.38 -> 39.58Hurt
raise pruning rate to p=99% (ablation)stochastic gradient pruning, ResNet-152 · CIFAR-10training accuracy (%): 95.70 -> 93.84Hurt
raise pruning rate to p=99% (ablation)stochastic gradient pruning, ResNet-152 · CIFAR-100training accuracy (%): 79.25 -> 76.40Hurt
raise pruning rate to p=99% (ablation)stochastic gradient pruning, ResNet-101 · CIFAR-100training accuracy (%): 79.70 -> 77.13Hurt
apply stochastic gradient pruning at p=70%ResNet-50 · CIFAR-100training accuracy (%): 77.74 -> 78.83Improved
apply stochastic gradient pruning at p=70%ResNet-18 · ImageNettraining accuracy (%): 68.73 -> 69.02No clear effect
apply stochastic gradient pruning at p=70%ResNet-34 · ImageNettraining accuracy (%): 72.93 -> 72.92No clear effect
apply stochastic gradient pruning at p=99%AlexNet · CIFAR-10training accuracy (%): 90.50 -> 89.66Hurt
apply stochastic gradient pruning at p=80%AlexNet · CIFAR-100training accuracy (%): 67.61 -> 68.13Improved
apply stochastic gradient pruning at p=99%ResNet-18 · CIFAR-10non-zero density: 1 -> 0.16Improved
use distribution-based threshold with stochastic pruning instead of MSBPResNet-18 · CIFAR-10non-zero density: 0.4 -> 0.16Improved
use distribution-based threshold with stochastic pruning instead of MSBPResNet-18 · CIFAR-10accuracy (%): 94.92 -> 95.18Improved

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