AI Paper Atlas

Research map / Architectures and efficiency / CNN design

Relay Backpropagation for Effective Learning of Deep Convolutional Neural Networks

Li Shen, Zhouchen Lin, Qingming Huang

ECCV 2016 · Architectures and efficiency › CNN design · approach: Filter design and trained-filter analysis · 464 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
use relay gradient limit instead of standard BPModel A · Places2 val (single crop)top-5 error: 19.00 -> 17.83Improved
use relay gradient limit instead of standard BPModel B · Places2 val (single crop)top-5 error: 18.69 -> 17.86Improved
add auxiliary loss with standard BP (multi-loss) (ablation)Model A · Places2 val (single crop)top-5 error: 19.00 -> 18.84No clear effect
add auxiliary loss with standard BP (multi-loss) (ablation)Model B · Places2 val (single crop)top-5 error: 18.69 -> 18.68No clear effect
limit each loss's gradient to its own segments (Relay BP) versus multi-loss with standard BP (ablation)Model A · Places2 val (single crop)top-5 error: 18.84 -> 17.83Improved
limit each loss's gradient to its own segments (Relay BP) versus multi-loss with standard BP (ablation)Model B · Places2 val (single crop)top-5 error: 18.68 -> 17.86Improved
use relay gradient limit instead of standard BPModel A · Places2 val (single model)top-5 error: 17.19 -> 16.33Improved
use relay gradient limit instead of standard BPModel B · Places2 val (single model)top-5 error: 16.89 -> 16.36Improved
use relay gradient limit instead of standard BPResNet-50 · ImageNet 2012 valtop-5 error: 5.37 -> 4.93Improved
use relay gradient limit instead of standard BPResNet-50 · ImageNet 2012 valtop-1 error: 21.17 -> 20.26Improved
use relay gradient limit instead of standard BPInception-v3 · ImageNet 2012 valtop-5 error: 4.43 -> 3.97Improved
use relay gradient limit instead of standard BPInception-v3 · ImageNet 2012 valtop-1 error: 19.18 -> 18.52Improved

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