Generalized multi-output Gaussian process censored regression

作者：

Highlights：

• Censored data as defining characteristic of numerous domains in science.

• Heteroscedastic Multi-Output Gaussian Process formulated for censored regression.

• Generalization of arbitrary likelihood functions enabled by devising a variational bound to the marginal log-likelihood.

• Experiments with synthetic and real-world data demonstrate solution approach.

摘要

•Censored data as defining characteristic of numerous domains in science.•Heteroscedastic Multi-Output Gaussian Process formulated for censored regression.•Generalization of arbitrary likelihood functions enabled by devising a variational bound to the marginal log-likelihood.•Experiments with synthetic and real-world data demonstrate solution approach.

论文关键词：Censored data,Gaussian processes,Variational inference

论文评审过程：Received 29 December 2020, Revised 8 April 2022, Accepted 25 April 2022, Available online 26 April 2022, Version of Record 2 May 2022.

论文官网地址：https://doi.org/10.1016/j.patcog.2022.108751