Deep learning with differential Gaussian process flows

Loading...
Thumbnail Image

Access rights

openAccess
acceptedVersion

URL

Journal Title

Journal ISSN

Volume Title

A4 Artikkeli konferenssijulkaisussa

Date

2019-04

Major/Subject

Mcode

Degree programme

Language

en

Pages

16

Series

The 22nd International Conference on Artificial Intelligence and Statistic, Volume 89, pp. 1-15, Proceedings of Machine Learning Research ; Volume 89

Abstract

We propose a novel deep learning paradigm of differential flows that learn a stochastic differential equation transformations of inputs prior to a standard classification or regression function. The key property of differential Gaussian processes is the warping of inputs through infinitely deep, but infinitesimal, differential fields, that generalise discrete layers into a dynamical system. We demonstrate state-of-the-art results that exceed the performance of deep Gaussian processes and neural networks

Description

Keywords

gaussian process, Bayesian methods

Other note

Citation

Hegde, P, Heinonen, M, Lähdesmäki, H & Kaski, S 2019, Deep learning with differential Gaussian process flows . in The 22nd International Conference on Artificial Intelligence and Statistic . vol. 89, Proceedings of Machine Learning Research, vol. 89, JMLR, pp. 1-15, International Conference on Artificial Intelligence and Statistics, Naha, Japan, 16/04/2019 . < https://proceedings.mlr.press/v89/hegde19a.html >