CUED Publications database

Deep Gaussian processes for regression using approximate expectation propagation

Bui, TD and Hernández-Lobato, JM and Hernández-Lobato, D and Li, Y and Turner, RE (2016) Deep Gaussian processes for regression using approximate expectation propagation. 33rd International Conference on Machine Learning, ICML 2016, 3. pp. 2187-2208.

Full text not available from this repository.

Abstract

Deep Gaussian processes (DGPs) are multilayer hierarchical generalisations of Gaussian processes (GPs) and are formally equivalent to neural networks with multiple, infinitely wide hidden layers. DGPs are nonparametric probabilistic models and as such are arguably more flexible, have a greater capacity to generalise, and provide better calibrated uncertainty estimates than alternative deep models. This paper develops a new approximate Bayesian learning scheme that enables DGPs to be applied to a range of medium to large scale regression problems for the first time. The new method uses an approximate Expectation Propagation procedure and a novel and efficient extension of the probabilistic backpropagation algorithm for learning. We evaluate the new method for non-linear regression on eleven real-world datasets, showing that it always outperforms GP regression and is almost always better than state-of-the-art deterministic and sampling-based approximate inference methods for Bayesian neural networks. As a by-product, this work provides a comprehensive analysis of six approximate Bayesian methods for training neural networks.

Item Type: Article
Subjects: UNSPECIFIED
Divisions: Div F > Computational and Biological Learning
Depositing User: Cron Job
Date Deposited: 17 Jul 2017 19:27
Last Modified: 03 Aug 2017 03:01
DOI: