The Continuous Formulation of Shallow Neural Networks as Wasserstein-Type Gradient Flows

DOI10.1007/978-3-031-05331-3_3OpenAlexW4312382631MaRDI QIDQ5886422

Xavier Fernández-Real, Alessio Figalli

Publication date: 5 April 2023

Published in: Analysis at Large (Search for Journal in Brave)

Full work available at URL: https://doi.org/10.1007/978-3-031-05331-3_3

zbMATH Keywords

gradient flowmachine learningWasserstein distancecontinuous formulation

Mathematics Subject Classification ID

Artificial neural networks and deep learning (68T07) Stochastic analysis (60Hxx) Transport equations (35Q49)

Cites Work

• Unnamed Item
• Unnamed Item
• Machine learning from a continuous viewpoint. I
• Analysis of a two-layer neural network via displacement convexity
• An invitation to optimal transport, Wasserstein distances, and gradient flows
• A mean-field optimal control formulation of deep learning
• A proposal on machine learning via dynamical systems
• The Variational Formulation of the Fokker--Planck Equation
• A mean field view of the landscape of two-layer neural networks
• Mean Field Analysis of Deep Neural Networks
• Mean Field Analysis of Neural Networks: A Law of Large Numbers

This page was built for publication: The Continuous Formulation of Shallow Neural Networks as Wasserstein-Type Gradient Flows