Deep learning with differential Gaussian process flows

Pashupati Hegde; Markus Heinonen; Harri Lähdesmäki; Samuel Kaski

Deep learning with differential Gaussian process flows

Pashupati Hegde
, Markus Heinonen
, Harri Lähdesmäki
, Samuel Kaski

Machine Learning and Robotics

Aalto University

Research output: Chapter in Book/Conference proceeding › Conference contribution › peer-review

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

Original language	English
Title of host publication	AISTATS 2019
Subtitle of host publication	22nd International Conference on Artificial Intelligence and Statistics
Publisher	Journal of Machine Learning Research
Pages	1812-1821
Number of pages	10
Volume	89
Publication status	Published - Apr 2019
Event	22nd International Conference on Artificial Intelligence and Statistics - Naha, Japan Duration: 16 Apr 2019 → 18 Apr 2019

Publication series

Name	Proceedings of Machine Learning Research
Publisher	MLResearchPress
ISSN (Electronic)	2640-3498

Conference

Conference	22nd International Conference on Artificial Intelligence and Statistics
Abbreviated title	AISTATS 2019
Country/Territory	Japan
City	Naha
Period	16/04/19 → 18/04/19

Keywords

gaussian process
Bayesian methods
Published in open access publication channel
International
113 Tietojenkäsittely ja informaatiotieteet
113 Computer and information sciences
113 Data- och informationsvetenskap

Cite this

@inproceedings{644e951257bc4409a82b2bff04bb5a40,

title = "Deep learning with differential Gaussian process flows",

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",

keywords = "gaussian process, Bayesian methods, Published in open access publication channel, International, 113 Tietojenk{\"a}sittely ja informaatiotieteet, 113 Computer and information sciences, 113 Data- och informationsvetenskap",

author = "Pashupati Hegde and Markus Heinonen and Harri L{\"a}hdesm{\"a}ki and Samuel Kaski",

year = "2019",

month = apr,

language = "English",

volume = "89",

series = "Proceedings of Machine Learning Research",

publisher = "Journal of Machine Learning Research",

pages = "1812--1821",

booktitle = "AISTATS 2019",

address = "United States",

note = "22nd International Conference on<br/>Artificial Intelligence and Statistics, AISTATS 2019 ; Conference date: 16-04-2019 Through 18-04-2019",

}

Hegde, P, Heinonen, M, Lähdesmäki, H & Kaski, S 2019, Deep learning with differential Gaussian process flows. in AISTATS 2019: 22nd International Conference on Artificial Intelligence and Statistics. vol. 89, Proceedings of Machine Learning Research, Journal of Machine Learning Research, pp. 1812-1821, 22nd International Conference on
Artificial Intelligence and Statistics, Naha, Japan, 16/04/19.

Deep learning with differential Gaussian process flows. / Hegde, Pashupati; Heinonen, Markus; Lähdesmäki, Harri et al.
AISTATS 2019: 22nd International Conference on Artificial Intelligence and Statistics. Vol. 89 Journal of Machine Learning Research, 2019. p. 1812-1821 (Proceedings of Machine Learning Research).

Research output: Chapter in Book/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Deep learning with differential Gaussian process flows

AU - Hegde, Pashupati

AU - Heinonen, Markus

AU - Lähdesmäki, Harri

AU - Kaski, Samuel

PY - 2019/4

Y1 - 2019/4

N2 - 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

AB - 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

KW - gaussian process

KW - Bayesian methods

KW - Published in open access publication channel

KW - International

KW - 113 Tietojenkäsittely ja informaatiotieteet

KW - 113 Computer and information sciences

KW - 113 Data- och informationsvetenskap

M3 - Conference contribution

VL - 89

T3 - Proceedings of Machine Learning Research

SP - 1812

EP - 1821

BT - AISTATS 2019

PB - Journal of Machine Learning Research

T2 - 22nd International Conference on<br/>Artificial Intelligence and Statistics

Y2 - 16 April 2019 through 18 April 2019

ER -

Deep learning with differential Gaussian process flows

Abstract

Publication series

Conference

Keywords

Fingerprint

Cite this