Gaussian conditional random fields for classification

作者：

Highlights：

• Gaussian conditional random field model for structured classification is proposed.

• Two different forms of the algorithm are presented Bayesian and non-Bayesian.

• The extension of local variational approximation of sigmoid function is presented.

• Both variants are evaluated on synthetic data and real-world data.

摘要

•Gaussian conditional random field model for structured classification is proposed.•Two different forms of the algorithm are presented Bayesian and non-Bayesian.•The extension of local variational approximation of sigmoid function is presented.•Both variants are evaluated on synthetic data and real-world data.

论文关键词：Structured classification,Gaussian conditional random fields,Empirical Bayes,Local variational approximation,Discriminative graph-based model

论文评审过程：Received 11 February 2020, Revised 11 August 2022, Accepted 28 August 2022, Available online 3 September 2022, Version of Record 19 September 2022.

论文官网地址：https://doi.org/10.1016/j.eswa.2022.118728