Healing Products of Gaussian Processes

02/14/2021
by   Samuel Cohen, et al.
0

Gaussian processes (GPs) are nonparametric Bayesian models that have been applied to regression and classification problems. One of the approaches to alleviate their cubic training cost is the use of local GP experts trained on subsets of the data. In particular, product-of-expert models combine the predictive distributions of local experts through a tractable product operation. While these expert models allow for massively distributed computation, their predictions typically suffer from erratic behaviour of the mean or uncalibrated uncertainty quantification. By calibrating predictions via a tempered softmax weighting, we provide a solution to these problems for multiple product-of-expert models, including the generalised product of experts and the robust Bayesian committee machine. Furthermore, we leverage the optimal transport literature and propose a new product-of-expert model that combines predictions of local experts by computing their Wasserstein barycenter, which can be applied to both regression and classification.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
02/02/2021

Gaussian Experts Selection using Graphical Models

Local approximations are popular methods to scale Gaussian processes (GP...
research
10/27/2020

Gaussian Model Trees for Traffic Imputation

Traffic congestion is one of the most pressing issues for smart cities. ...
research
05/07/2017

Learning of Gaussian Processes in Distributed and Communication Limited Systems

It is of fundamental importance to find algorithms obtaining optimal per...
research
11/17/2022

Expert Selection in Distributed Gaussian Processes: A Multi-label Classification Approach

By distributing the training process, local approximation reduces the co...
research
02/27/2019

Local Bandwidth Estimation via Mixture of Gaussian Processes

Real world data often exhibit inhomogeneity - complexity of the target f...
research
10/28/2014

Generalized Product of Experts for Automatic and Principled Fusion of Gaussian Process Predictions

In this work, we propose a generalized product of experts (gPoE) framewo...
research
06/16/2021

Leveraging Probabilistic Circuits for Nonparametric Multi-Output Regression

Inspired by recent advances in the field of expert-based approximations ...

Please sign up or login with your details

Forgot password? Click here to reset