Manifold structured prediction
File(s)
Published version
OA Location
Author(s)
Rudi, Alessandro
Ciliberto, Carlo
Marconi, Gian Maria
Rosasco, Lorenzo
Type
Conference Paper
Abstract
Structured prediction provides a general framework to deal with supervised problems where the outputs have semantically rich structure. While classical approaches consider finite, albeit potentially huge, output spaces, in this paper we discuss how structured prediction can be extended to a continuous scenario. Specifically, we study a structured prediction approach to manifold-valued regression. We characterize a class of problems for which the considered approach is statistically consistent and study how geometric optimization can be used to compute the corresponding estimator. Promising experimental results on both simulated and real data complete our study.
Editor(s)
Bengio, S
Wallach, H
Larochelle, H
Grauman, K
CesaBianchi, N
Garnett, R
Date Issued
2018-12-02
Date Acceptance
2018-12-02
Citation
Advances in Neural Information Processing Systems, 2018, 31
ISSN
1049-5258
Publisher
Massachusetts Institute of Technology Press
Journal / Book Title
Advances in Neural Information Processing Systems
Volume
31
Copyright Statement
© 2018 Massachusetts Institute of Technology Press.
Identifier
http://gateway.webofknowledge.com/gateway/Gateway.cgi?GWVersion=2&SrcApp=PARTNER_APP&SrcAuth=LinksAMR&KeyUT=WOS:000461852000014&DestLinkType=FullRecord&DestApp=ALL_WOS&UsrCustomerID=1ba7043ffcc86c417c072aa74d649202
Source
32nd Conference on Neural Information Processing Systems (NIPS)
Subjects
Science & Technology
Technology
Computer Science, Artificial Intelligence
Computer Science
LEAST-SQUARES
REGRESSION
Publication Status
Published
Start Date
2018-12-02
Finish Date
2018-12-08
Coverage Spatial
Montreal, Canada
Date Publish Online
2018-12-02