AI Paper Atlas

Research map / Architectures and efficiency / RNNs and normalization

Deep Polynomial Neural Networks

Grigorios G. Chrysos, Stylianos Moschoglou, Giorgos Bouritsas, Jiankang Deng, Yannis Panagakis, Stefanos P. Zafeiriou

TPAMI 2021 · Architectures and efficiency › RNNs and normalization · approach: Theory of depth and width · 107 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
convert SNGAN generator to single NCP polynomialSNGAN · CIFAR10 generationFID: 19.06 -> 17.65Improved
use product of polynomials instead of single NCP (ablation)SNGAN generator · CIFAR10 generationFID: 17.65 -> 16.79Improved
convert StyleGAN mapping network to NCP polynomialStyleGAN · FFHQ 256x256FID: 7.15 -> 6.82Improved
use second-order residual blocks to cut blocks of ResNet34ResNet34 · Speech Commandsaccuracy: 0.951 -> 0.951, params 21.3 -> 13.2No clear effect
second-order residual blocks with fewer blocksResNet18 · CIFAR10accuracy: 0.945 -> 0.945, params 11.2 -> 6.0No clear effect
second-order residual blocks with fewer blocksResNet34 · CIFAR100accuracy: 0.769 -> 0.769, params 21.3 -> 14.7No clear effect
second-order residual blocks (NCP-Skip) in ResNet50ResNet50 · ImageNetTop-1 error: 23.546 -> 22.827Improved
second-order residual blocks in ResNet50ResNet50 · ImageNet trainingthroughput: 1625 -> 1531Hurt
second-order residual blocks in ArcFace ResNet50ResNet50 · IJB-B 1:1 verification FAR=1e-6TAR: 37.28 -> 43.46Improved
second-order residual blocks in ArcFace ResNet50ResNet50 · CPLFWverification accuracy: 92.433 -> 93.317Improved
second-order residual blocks in ArcFace ResNet50ResNet50 · MegaFaceId / Ver: 98.28 -> 98.78 (Id); 98.64 -> 98.95 (Ver)Improved
increase polynomial order of Spiral Pi-Net, simple model (ablation)Spiral mesh autoencoder · COMA and DFAUST mesh reconstructionreconstruction errorMixed

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