AI Paper Atlas

Research map / Optimization / Implicit bias and generalization

Complex Dynamics in Simple Neural Networks: Understanding Gradient Flow in Phase Retrieval

Stefano Sarao Mannelli, Giulio Biroli, Chiara Cammarota, Florent Krzakala, Pierfrancesco Urbani, Lenka Zdeborová

NeurIPS 2020 · Optimization › Implicit bias and generalization · approach: Gradient flow and mean-field dynamics · 18 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
Increase the sample-to-dimension ratio above the empirical recovery regimeRandomly initialized gradient descent for phase retrieval · Phase retrievalzero test error: before α∼7.0Improved
Increase the sample-to-dimension ratio from α=6 to α=18 or α=24Gradient flow for phase retrieval · Phase retrievaltraining loss convergence to zero: α=6; α=18; 24Improved
Operate near the critical sample-to-dimension ratioGradient flow for phase retrieval · Phase retrieval, α=12perfect generalization across runs: α=12Mixed
Use the threshold-state Hessian's signal-aligned unstable direction after the BBP transitionGradient flow for phase retrieval · Phase retrieval, above the critical ratiorecovery of the global minimumImproved

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