Learning performance of regularized regression with multiscale kernels based on Markov observations

作者：

Highlights：

• extend the multiscale kernels learning algorithm with the i.i.d. observations to the u.e.M.c. observations.

• propose a novel Markov observing algorithm with the MLSRR scheme to generate u.e.M.c. samples.

• MLSRR-M has better generalization performance compared to that of i.i.d. observations.

摘要

•extend the multiscale kernels learning algorithm with the i.i.d. observations to the u.e.M.c. observations.•propose a novel Markov observing algorithm with the MLSRR scheme to generate u.e.M.c. samples.•MLSRR-M has better generalization performance compared to that of i.i.d. observations.

论文关键词：Learning performance,Uniformly ergodic Markov chain (u.e.M.c.),Markov observations,Multiscale kernels

论文评审过程：Received 2 June 2020, Revised 11 May 2021, Accepted 16 May 2021, Available online 12 June 2021, Version of Record 12 June 2021.

论文官网地址：https://doi.org/10.1016/j.amc.2021.126386