Modulating scalable Gaussian processes for expressive statistical learning
作者：
Highlights：
• New scalable Gaussian process (GP) paradigms to introduce additional modulation variables for learning rich statistical representation, e.g., het- eroscedastic noise, multi-modality and non-stationarity, from massive data.
• Different variational inference strategies to arrive at analytical or tight evidence lower bounds (ELBOs) for effcient and effective model training.
• Comprehensive comparison against state-of-the-art GP and neural net- work counterparts to showcase the superiority of scalable modulated GPs.
摘要
•New scalable Gaussian process (GP) paradigms to introduce additional modulation variables for learning rich statistical representation, e.g., het- eroscedastic noise, multi-modality and non-stationarity, from massive data.•Different variational inference strategies to arrive at analytical or tight evidence lower bounds (ELBOs) for effcient and effective model training.•Comprehensive comparison against state-of-the-art GP and neural net- work counterparts to showcase the superiority of scalable modulated GPs.
论文关键词：Gaussian process,Modulation,Scalability,Heteroscedastic noise,Multi-modality,Non-stationarity
论文评审过程：Received 29 August 2020, Revised 15 June 2021, Accepted 19 June 2021, Available online 2 July 2021, Version of Record 2 July 2021.
论文官网地址：https://doi.org/10.1016/j.patcog.2021.108121
