English
 
Help Privacy Policy Disclaimer
  Advanced SearchBrowse

Item

ITEM ACTIONSEXPORT
Add to Basket
 Local TagsRelease HistoryDetailsSummary
  It is all in the noise: Efficient multi-task Gaussian process inference with structured residuals

Rakitsch, B., Lippert, C., Borgwardt, K., & Stegle, O. (2014). It is all in the noise: Efficient multi-task Gaussian process inference with structured residuals. In C. Burges, L. Bottou, M. Welling, Z. Ghahramani, & K. Weinberger (Eds.), Advances in Neural Information Processing Systems 26 (pp. 1468-1476). Red Hook, NY, USA: Curran.

Item is

 

show all hide all

Basic

show hide
Item Permalink: https://hdl.handle.net/21.11116/0000-0010-5B0B-D Version Permalink: https://hdl.handle.net/21.11116/0000-0010-7A72-5
Genre: Conference Paper

Files

show Files

Locators

show
hide
Locator:
https://proceedings.neurips.cc/paper/2013/hash/59c33016884a62116be975a9bb8257e3-Abstract.html (Publisher version)
Description:
-
OA-Status:
Not specified

Creators

show
hide
 Creators:
Rakitsch, B1, Author                 
Lippert, C, Author                 
Borgwardt, K1, Author                 
Stegle, O, Author                 
Affiliations:
1Department Molecular Biology, Max Planck Institute for Developmental Biology, Max Planck Society, ou_3375790              

Content

show
hide
Free keywords: -
 Abstract: Multi-task prediction models are widely being used to couple regressors or classification models by sharing information across related tasks. A common pitfall of these models is that they assume that the output tasks are independent conditioned on the inputs. Here, we propose a multi-task Gaussian process approach to model both the relatedness between regressors as well as the task correlations in the residuals, in order to more accurately identify true sharing between regressors. The resulting Gaussian model has a covariance term that is the sum of Kronecker products, for which efficient parameter inference and out of sample prediction are feasible. On both synthetic examples and applications to phenotype prediction in genetics, we find substantial benefits of modeling structured noise compared to established alternatives.

Details

show
hide
Language(s): eng - English
 Dates: 2014-04
 Publication Status: Issued
 Pages: -
 Publishing info: -
 Table of Contents: -
 Rev. Type: -
 Identifiers: -
 Degree: -

Event

show
hide
Title: 27th Annual Conference on Neural Information Processing Systems (NIPS 2013)
Place of Event: Lake Tahoe, NV, USA
Start-/End Date: 2013-12-05 - 2013-12-10

Legal Case

show

Project information

show

Source 1

show
hide
Title: Advances in Neural Information Processing Systems 26
Source Genre: Proceedings
 Creator(s):
Burges, CJ1, Editor
Bottou, L1, Editor
Welling, M1, Editor
Ghahramani, Z1, Editor
Weinberger, KQ1, Editor
Affiliations:
1 External Organizations, ou_persistent22            
Publ. Info: Red Hook, NY, USA : Curran
Pages: - Volume / Issue: 2 Sequence Number: - Start / End Page: 1468 - 1476 Identifier: ISBN: 978-1-63266-024-4