AI Paper Atlas

Research map / Architectures and efficiency / RNNs and normalization

Momentum Residual Neural Networks

Michael E. Sander, Pierre Ablin, Mathieu Blondel, Gabriel Peyré

ICML 2021 · Architectures and efficiency › RNNs and normalization · approach: Efficient network compression and training · 10 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
Replace ResNet blocks with Momentum ResNet blocks using v₀ = 0ResNet-101 · CIFAR-10test accuracy: 95.15 ± 0.12 -> 95.1 ± 0.13No clear effect
Replace ResNet blocks with Momentum ResNet blocks using v₀ = 0ResNet-101 · CIFAR-100test accuracy: 76.86 ± 0.25 -> 76.39 ± 0.18Hurt
Use v₀ = f(x₀) instead of v₀ = 0 (ablation)Momentum ResNet-101 · CIFAR-10test accuracy: 95.1 ± 0.13 -> 95.18 ± 0.06No clear effect
Use v₀ = f(x₀) instead of v₀ = 0 (ablation)Momentum ResNet-101 · CIFAR-100test accuracy: 76.39 ± 0.18 -> 76.38 ± 0.42No clear effect
Vary the momentum parameter γ (ablation)Momentum ResNet-101 · CIFAR-10test accuracy: never decreases by more than 3 %No clear effect
Learn γ instead of fixing it (ablation)Momentum ResNet-101 · CIFAR-10test accuracyNo clear effect
Transfer pretrained ResNet-152 parameters to Momentum ResNet-152ImageNet-pretrained ResNet-152 · ImageNettop-1 error: 26.5 %; 23.5 %Improved
Use Momentum ResNet-152 rather than ResNet-152 for fine-tuningImageNet-pretrained models · hymenopterafine-tuning convergence speed: twice as fastImproved
Add momentum to first-order ODE blocksCIFAR-10 ODE network · CIFAR-10test accuracyImproved
Use Momentum ResNet dynamics instead of RevNet dynamicsLearned-ISTA · Lasso sparse coding test examplestest lossImproved
Use Momentum ResNet instead of LISTALearned-ISTA · Lasso sparse coding test examplestest lossImproved
Use Momentum ResNet rather than ResNet for separating nested ringsDepth-15 residual network · Four nested ringsclass separability: 4Improved

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