AI Paper Atlas

Research map / Architectures and efficiency / RNNs and normalization

SP-Net: Slowly Progressing Dynamic Inference Networks

Huanyu Wang, Wenhu Zhang, Shihao Su, Hui Wang, Zhenwei Miao, Xin Zhan, Xi Li

ECCV 2022 · Architectures and efficiency › RNNs and normalization · approach: Efficient network compression and training · 3 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
add the dynamic auxiliary module (momentum teacher with path and logits distillation) (ablation)vanilla dynamic inference, ResNet-110 · CIFAR-10top-1 accuracy (%): 93.83 -> 94.40Improved
add the dynamic auxiliary module (ablation)vanilla dynamic inference, ResNet-110 · CIFAR-100top-1 accuracy (%): 70.17 -> 73.63Improved
add multi-stage classification (ablation)SP-Net with auxiliary module, ResNet-110 · CIFAR-10top-1 accuracy (%): 94.40 -> 95.04Improved
add multi-stage classification (ablation)SP-Net with auxiliary module, ResNet-110 · CIFAR-100top-1 accuracy (%): 73.63 -> 74.91Improved
add feature evolution direction regularization (ablation)SP-Net A+B, ResNet-110 · CIFAR-10top-1 accuracy (%): 95.04 -> 95.22Improved
add feature evolution direction regularization (ablation)SP-Net A+B, ResNet-110 · CIFAR-100top-1 accuracy (%): 74.91 -> 75.14Improved
raise momentum coefficient from 0.7 to 0.8 (ablation)SP-Net, ResNet-110 · CIFAR-10top-1 accuracy (%): 95.22 -> 94.82Hurt
lower momentum coefficient from 0.7 to 0.5 (ablation)SP-Net, ResNet-110 · CIFAR-10top-1 accuracy (%): 95.22 -> 94.97Hurt
regularize feature evolution across different resolutions instead of all adjacent blocks (ablation)SP-Net, ResNet-110 · CIFAR-10top-1 accuracy (%): 94.43 -> 95.22Improved
regularize feature evolution across resolutions instead of adjacent blocks in same resolution (ablation)SP-Net, ResNet-110 · CIFAR-100top-1 accuracy (%): 74.05 -> 74.86Improved
use SP-Net instead of vanilla dynamic baseline at target executing rate 0.2dynamic ResNet-110 · CIFAR-10top-1 accuracy (%): 90.48 with 0.11 GMACCs -> 93.75 at the computational cost of 0.14 GMACCsImproved
use SP-Net instead of vanilla dynamic baseline at target executing rate 0.4dynamic ResNet-110 · CIFAR-10top-1 accuracy (%): 93.52 -> 94.47Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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