Current browse context:
math.AP
Change to browse by:
References & Citations
Mathematics > Analysis of PDEs
Title: Convergence rate of DeepONets for learning operators arising from advection-diffusion equations
(Submitted on 21 Feb 2021 (v1), last revised 17 Mar 2021 (this version, v2))
Abstract: We present convergence analysis of operator learning in [Chen and Chen 1995] and [Lu et al. 2020], where continuous operators are approximated by a sum of products of branch and trunk networks. In this work, we consider the rates of learning solution operators from both linear and nonlinear advection-diffusion equations with or without reaction. We find that the convergence rates depend on the architecture of branch networks as well as the smoothness of inputs and outputs of solution operators.
Submission history
From: Zhongqiang Zhang [view email][v1] Sun, 21 Feb 2021 14:55:39 GMT (26kb)
[v2] Wed, 17 Mar 2021 15:14:13 GMT (400kb,D)
Link back to: arXiv, form interface, contact.