Change-point detection in hierarchical circadian models
作者：
Highlights：
• Bayesian change-point detection on sequences of high-dimensional and heterogeneous observations with temporal structure.
• Heterogeneous-cirdadian mixture models with non-stationary and periodic covariance functions.
• Maximum-a-posteriori (MAP) detection from low-dimensional time-series of discrete latent variables.
• Applied experiments to human behavior modelling and detection of changes in mental health patients.
摘要
•Bayesian change-point detection on sequences of high-dimensional and heterogeneous observations with temporal structure.•Heterogeneous-cirdadian mixture models with non-stationary and periodic covariance functions.•Maximum-a-posteriori (MAP) detection from low-dimensional time-series of discrete latent variables.•Applied experiments to human behavior modelling and detection of changes in mental health patients.
论文关键词：Change-point detection,Circadian models,Heterogeneous data,Latent variable models,Non-stationary periodic covariance functions
论文评审过程：Received 23 January 2020, Revised 15 August 2020, Accepted 26 December 2020, Available online 13 January 2021, Version of Record 2 February 2021.
论文官网地址：https://doi.org/10.1016/j.patcog.2021.107820
