New theory uses neural networks for fractional parabolic equations

Researchers have developed a new neural network approximation theory for solving fractional parabolic partial differential equations. This theory utilizes anisotropic spectral Barron spaces to analyze temporal and spatial regularity separately in frequency space. The approach incorporates lower-order drift and potential terms through dimension-independent multiplication estimates and the method of continuity. The study also explores the global-in-time extension of fractional heat semigroups using Vandermonde matrices and derives approximation bounds for various activation functions. AI

IMPACT This research could lead to more efficient neural network models for solving complex differential equations in scientific computing.

RANK_REASON The cluster contains a single academic paper detailing a new mathematical theory and approximation method. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.LG →

• alphaXiv
• Anisotropic Spectral Barron Spaces
• arXiv
• CatalyzeX Code Finder for Papers
• Connected Papers
• CORE Recommender
• DagsHub
• Fractional heat semigroups on metric measure spaces with finite densities and applications to fractional dissipative equations
• Gotit.pub
• Hugging Face
• Influence Flower
• Litmaps
• Mixed Sobolev Norms
• ScienceCast
• scite Smart Citations
• Vandermonde matrix

• paper
• other

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →