Distributed robust Gaussian Process regression

Research output: Journal contributionsJournal articlesResearchpeer-review

Standard

Distributed robust Gaussian Process regression. / Mair, Sebastian; Brefeld, Ulf.
In: Knowledge and Information Systems, Vol. 55, No. 2, 01.05.2018, p. 415-435.

Research output: Journal contributionsJournal articlesResearchpeer-review

Harvard

APA

Vancouver

Mair S, Brefeld U. Distributed robust Gaussian Process regression. Knowledge and Information Systems. 2018 May 1;55(2):415-435. Epub 2017 Jul 19. doi: 10.1007/s10115-017-1084-7

Bibtex

@article{e7e0e3faacc7473b99d76b870a3b77d6,
title = "Distributed robust Gaussian Process regression",
abstract = "We study distributed and robust Gaussian Processes where robustness is introduced by a Gaussian Process prior on the function values combined with a Student-t likelihood. The posterior distribution is approximated by a Laplace Approximation, and together with concepts from Bayesian Committee Machines, we efficiently distribute the computations and render robust GPs on huge data sets feasible. We provide a detailed derivation and report on empirical results. Our findings on real and artificial data show that our approach outperforms existing baselines in the presence of outliers by using all available data.",
keywords = "Business informatics, Distributed computation, Gaussian process regression, Laplace Approximation, Robust regression, Student-t likelihood",
author = "Sebastian Mair and Ulf Brefeld",
year = "2018",
month = may,
day = "1",
doi = "10.1007/s10115-017-1084-7",
language = "English",
volume = "55",
pages = "415--435",
journal = "Knowledge and Information Systems",
issn = "0219-1377",
publisher = "Springer London",
number = "2",

}

RIS

TY - JOUR

T1 - Distributed robust Gaussian Process regression

AU - Mair, Sebastian

AU - Brefeld, Ulf

PY - 2018/5/1

Y1 - 2018/5/1

N2 - We study distributed and robust Gaussian Processes where robustness is introduced by a Gaussian Process prior on the function values combined with a Student-t likelihood. The posterior distribution is approximated by a Laplace Approximation, and together with concepts from Bayesian Committee Machines, we efficiently distribute the computations and render robust GPs on huge data sets feasible. We provide a detailed derivation and report on empirical results. Our findings on real and artificial data show that our approach outperforms existing baselines in the presence of outliers by using all available data.

AB - We study distributed and robust Gaussian Processes where robustness is introduced by a Gaussian Process prior on the function values combined with a Student-t likelihood. The posterior distribution is approximated by a Laplace Approximation, and together with concepts from Bayesian Committee Machines, we efficiently distribute the computations and render robust GPs on huge data sets feasible. We provide a detailed derivation and report on empirical results. Our findings on real and artificial data show that our approach outperforms existing baselines in the presence of outliers by using all available data.

KW - Business informatics

KW - Distributed computation

KW - Gaussian process regression

KW - Laplace Approximation

KW - Robust regression

KW - Student-t likelihood

UR - http://www.scopus.com/inward/record.url?scp=85025086950&partnerID=8YFLogxK

U2 - 10.1007/s10115-017-1084-7

DO - 10.1007/s10115-017-1084-7

M3 - Journal articles

VL - 55

SP - 415

EP - 435

JO - Knowledge and Information Systems

JF - Knowledge and Information Systems

SN - 0219-1377

IS - 2

ER -

Recently viewed

Activities

  1. Interstitial spaces as garbage cans of field transformation where problems and solutions meet: the case of blockchain and music cross-fertilization
  2. Big Data, Data Trading, and Data Protection
  3. A utilitarian notion of responsibility for sustainability
  4. Responding ASAP? The Role of Age in Dealing with Availability and Response Expectations
  5. Eine Podiumsdiskussion zu Fracking
  6. Negotiating normativity: discourses of (non) belonging and (non) coincidences in the context of transnational adoption
  7. German Teaching and Learning Materials - Lifelong Learning and Competency-Based Instruction
  8. 'Can you play a new CD,please?' Speech act representation in EFL textbooks: An interlanguage pragmatic appraisal (Universität Gießen, invited talk)
  9. What if Civilization Collapses? Management Scholarship in and for Deep Adaption
  10. Can we solve the climate crisis? Contributions from artS, technology and science
  11. WK ORG Workshop - WK ORG 2019
  12. How to make use of evidence based management in entrepreneurship: The example of Personal Initiative Training
  13. Effective working hours and wages: the case of downward adjustment via paid absenteeism
  14. Collective Decisions
  15. “Visual Rhetoric as a three-dimensional practice”
  16. Licht im Dunkeln 2005
  17. 5th European Conference of Apidology - EurBee 2012
  18. Reactivity on a student-centered recording system – a video-based longitudinal study in primary education
  19. Tolerating and inducing temporal asynchronicity in complex innovation journeys
  20. Personal care products as source for micropollutants in Greywater-Identification, quantification and on-site treatment
  21. Sustainability Reporting on the World Wide Web: Developments in Germany
  22. Applied Vegetation Science (Zeitschrift)
  23. Universität Zürich
  24. Deferred Compensation Schemes, Fairness Concerns, and Employment of Older Workers
  25. Forschungspraxis und Selbstsorge in Sensitive Research
  26. Biological Oxidation of Iron with Various Oxidants.
  27. Purpuseful Work Symposium