AI Paper Atlas

Research map / Architectures and efficiency / RNNs and normalization

Hierarchical nucleation in deep neural networks

Diego Doimo, Aldo Glielmo, Alessio Ansuini, Alessandro Laio

NeurIPS 2020 · Architectures and efficiency › RNNs and normalization · approach: Representation geometry and collapse · 2 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
compare ImageNet with MNIST, modMNIST and CIFAR-10 for the sharpness of the chi(l,gt) transitionResNet152 · ILSVRC2012 vs MNIST/CIFAR-10chi(l,gt) transition sharpnessMixed
increase neighborhood size k to 300 (ablation)chi(l,gt) and chi(l,out) profiles · ResNet152, ILSVRC2012 subsettransition detectability: k=300Hurt
vary dataset size N with constant class ratio (ablation)chi(l,gt) and chi(l,out) profiles · ResNet152, ILSVRC2012 subsetchi profileNo clear effect
decrease Gaussian CKA kernel width sigma (ablation)Gaussian CKA with output layer · ResNet152, 5000 ILSVRC2012 imagestransition sharpnessImproved
use linear CKA instead of neighborhood overlap (ablation)similarity measure · ResNet152, 5000 ILSVRC2012 imagesdetection of abrupt transitionHurt
change architecture across ResNet and VGG depthschi(l,gt) profile · ILSVRC2012 subsetchi(l,gt) trendNo clear effect
fewer classes (10) in modMNIST and CIFAR-10 versus 300 in ImageNetResNet152 · modMNIST, CIFAR-10lag phase of chi(l,gt)Improved

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