Is SGD a Bayesian sampler? Well, almost

MaRDI QIDQ4998999zbMATH OpenFDO

Authors Chris Mingard, Guillermo Valle-Pérez, Joar Skalse, Ard A. Louis

Publication date 9 July 2021

Full work available at URL https://arxiv.org/abs/2006.15191, https://jmlr.csail.mit.edu/papers/v22/20-676.html

Mathematics Subject Classification ID

Learning and adaptive systems in artificial intelligence (68T05)

Recommendations

• Dynamics of stochastic gradient descent for two-layer neural networks in the teacher–student setup*
• Natural Langevin dynamics for neural networks
• scientific article; zbMATH DE number 6860839
• A mean field view of the landscape of two-layer neural networks
• Over-parametrized deep neural networks minimizing the empirical risk do not generalize well

Cited in

(2)

• Estimating time-varying reproduction number by deep learning techniques
• Generalization performance of narrow shallow neural networks in the teacher-student setting

This page was built for publication: Is SGD a Bayesian sampler? Well, almost

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4998999)