Bayesian Alignments of Warped Multi-Output Gaussian Processes

Markus Kaiser, Clemens Otte, Thomas Runkler, Carl Henrik Ek

Research output: Chapter in Book/Report/Conference proceedingConference Contribution (Conference Proceeding)

96 Downloads (Pure)


We propose a novel Bayesian approach to modelling nonlinear alignments of time series based on latent shared information. We apply the method to the real-world problem of finding common structure in the sensor data of wind turbines introduced by the underlying latent and turbulent wind field. The proposed model allows for both arbitrary alignments of the inputs and non-parametric output warpings to transform the observations. This gives rise to multiple deep Gaussian process models connected via latent generating processes. We present an efficient variational approximation based on nested variational compression and show how the model can be used to extract shared information between dependent time series, recovering an interpretable functional decomposition of the learning problem. We show results for an artificial data set and real-world data of two wind turbines.
Original languageEnglish
Title of host publicationAdvances in Neural Information Processing Systems 31
Subtitle of host publicationAnnual Conference on Neural Information Processing Systems 2018, NeurIPS 2018, 3-8 December 2018, Montréal, Canada.
EditorsSamy Bengio, Hanna M. Wallach, Hugo Larochelle, Kristen Grauman, Nicolò Cesa-Bianchi, Roman Garnett
PublisherAssociation for Computing Machinery (ACM)
Number of pages10
Publication statusPublished - 3 Dec 2018

Publication series



  • stat.ML
  • cs.LG


Dive into the research topics of 'Bayesian Alignments of Warped Multi-Output Gaussian Processes'. Together they form a unique fingerprint.

Cite this