Timezone: »
We identify a new variational inference scheme for dynamical systems whose transition function is modelled by a Gaussian process. Inference in this setting has, so far, either employed computationally intensive MCMC methods, or relied on factorisations of the variational posterior. As we demonstrate in our experiments, the factorisation between latent system states and transition function can lead to a miscalibrated posterior and to learning unnecessarily large noise terms. We eliminate this factorisation by explicitly modelling the dependence between the states and the low-rank representation of our Gaussian process posterior. Samples of the latent states can then be tractably generated by conditioning on this representation. The method we obtain gives better predictive performance and more calibrated estimates of the transition function, yet maintains the same time and space complexities as mean-field methods.
Author Information
Alessandro Davide Ialongo (University of Cambirdge; Max Planck Tübingen)
Mark van der Wilk (PROWLER.io)
James Hensman (PROWLER.io)
Carl E Rasmussen (Cambridge University)
Related Events (a corresponding poster, oral, or spotlight)
-
2019 Poster: Overcoming Mean-Field Approximations in Recurrent Gaussian Process Models »
Thu. Jun 13th 01:30 -- 04:00 AM Room Pacific Ballroom #215
More from the Same Authors
-
2022 : Contrasting Discrete and Continuous Time Methods for Bayesian System Identification »
Talay Cheema · Carl E Rasmussen -
2019 Oral: Rates of Convergence for Sparse Variational Gaussian Process Regression »
David Burt · Carl E Rasmussen · Mark van der Wilk -
2019 Poster: Deep Gaussian Processes with Importance-Weighted Variational Inference »
Hugh Salimbeni · Vincent Dutordoir · James Hensman · Marc P Deisenroth -
2019 Oral: Deep Gaussian Processes with Importance-Weighted Variational Inference »
Hugh Salimbeni · Vincent Dutordoir · James Hensman · Marc P Deisenroth -
2019 Poster: Rates of Convergence for Sparse Variational Gaussian Process Regression »
David Burt · Carl E Rasmussen · Mark van der Wilk -
2018 Poster: Large-Scale Cox Process Inference using Variational Fourier Features »
ST John · James Hensman -
2018 Oral: Large-Scale Cox Process Inference using Variational Fourier Features »
ST John · James Hensman -
2018 Poster: PIPPS: Flexible Model-Based Policy Search Robust to the Curse of Chaos »
Paavo Parmas · Carl E Rasmussen · Jan Peters · Kenji Doya -
2018 Oral: PIPPS: Flexible Model-Based Policy Search Robust to the Curse of Chaos »
Paavo Parmas · Carl E Rasmussen · Jan Peters · Kenji Doya