Warped Gaussian Processes 论文

摘要

We generalise the Gaussian process (GP) framework for regression by\nlearning a nonlinear transformation of the GP outputs. This allows for\nnon-Gaussian processes and non-Gaussian noise. The learning algorithm\nchooses a nonlinear transformation such that transformed data is\nwell-modelled by a GP. This can be seen as including a preprocessing\ntransformation as an integral part of the probabilistic modelling\nproblem, rather than as an ad-hoc step. We demonstrate on several real\nregression problems that learning the transformation can lead to\nsignificantly better performance than using a regular GP, or a GP with\na fixed transformation.