Adaptive LOOCV-based kernel methods for solving time-dependent BVPs

作者：

Highlights：

• An adaptive algorithm is implemented for solving time-dependent boundary value problems.

• The adaptive scheme is based on a leave-one-out cross validation error indicator.

• The point distribution is refined by applying a two-point refinement strategy.

• Numerical experiments show efficacy and performance of the new adaptive method.

摘要

•An adaptive algorithm is implemented for solving time-dependent boundary value problems.•The adaptive scheme is based on a leave-one-out cross validation error indicator.•The point distribution is refined by applying a two-point refinement strategy.•Numerical experiments show efficacy and performance of the new adaptive method.

论文关键词：Meshless approximation,Kernel methods,Adaptive algorithms,Refinement schemes,Partial differential equations

论文评审过程：Received 25 February 2022, Revised 30 April 2022, Accepted 4 May 2022, Available online 14 May 2022, Version of Record 14 May 2022.

论文官网地址：https://doi.org/10.1016/j.amc.2022.127228