AI Paper Atlas

Research map / Optimization / Implicit bias and generalization

Generalization Bounds for Noisy Iterative Algorithms Using Properties of Additive Noise Channels

Hao Wang, Rui Gao, Flavio P. Calmon

JMLR 2023 · Optimization › Implicit bias and generalization · approach: SGD noise and flat minima · 9 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 the variance-based generalization bound instead of the gradient-prediction-residual boundgeneralization bound · MNISTKendall's tau: 0.25 -> 0.50Improved
Use the variance-based generalization bound instead of the gradient-prediction-residual boundgeneralization bound · CIFAR-10Kendall's tau: 0.38 -> 0.45Improved
Use the variance-based bound instead of the baseline bound, conditional independence testgeneralization bound · MNISTMI: 0.12 -> 0.34Improved
Use the variance-based bound instead of the baseline bound, varying network widthgeneralization bound · MNISTcorrelation, width varied: 0.26 -> 1.00Improved
Increase label corruption level from 0% to 75% of training samplesSGLD-trained network · corrupted labelsgeneralization bound and gap: bound and gap both increase with corruption levelHurt
Increase network widthSGLD-trained network · network width experimentgeneralization gap and bound: both decrease with widthImproved

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