AI Paper Atlas

Research map / Optimization / Implicit bias and generalization

Imitating Deep Learning Dynamics via Locally Elastic Stochastic Differential Equations

Jiayao Zhang, Hua Wang, Weijie J Su

NeurIPS 2021 · Optimization › Implicit bias and generalization · approach: Edge-of-stability dynamics · 1 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 logits-as-features (L-model) instead of isotropic features (I-model) to estimate dynamicsLE-ODE simulations on GeoMNIST and CIFAR · GeoMNIST and CIFAR, K=3logit-dynamics fidelityImproved
Use isotropic-feature-based estimates instead of logits-as-features estimates (ablation)LE-ODE simulations on GeoMNIST and CIFAR · GeoMNIST and CIFAR, K=3logit-dynamics fidelityHurt
Increase label corruption toward random labelingAlexNet training · GeoMNISTvalidation accuracy: 2/3Hurt
Increase label corruption toward random labelingAlexNet training · GeoMNISTfeature separability: around p = 0.6Hurt
Make intra-class elasticity greater than inter-class elasticityLE-SDE · Binary classificationasymptotic feature separability: α > βImproved
Set intra-class elasticity equal to or below inter-class elasticityLE-SDE · Binary classificationasymptotic feature separability: α ≤ βHurt

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