A Probabilistic Model for Dirty Multi-task Feature Selection
Entity
UAM. Departamento de Ingeniería InformáticaPublisher
International Machine Learning Society (IMLS)Date
2015Citation
32nd International Conference on Machine Learning, ICML 2015. JMLR, 2015. 1073-1082ISSN
1938-7228ISBN
978-151081058-7Funded by
Daniel Hernández-Lobato gratefully acknowledges the use of the facilities of Centro de Computacin Científica (CCC) at Universidad Autónoma de Madrid. This author also acknowledges financial support from Spanish Plan Nacional I+D+i, Grant TIN2013-42351-P, and from Comunidad de Madrid, Grant S2013/ICE-2845 CASI-CAM-CM. José Miguel Hernández-Lobato acknowledges financial support from the Rafael del Pino Fundation.Project
Gobierno de España. TIN2013-42351-P; Comunidad de Madrid. S2013/ICE-2845/CASI-CAMEditor's Version
http://www.jmlr.org/proceedings/papers/v37/Subjects
Artificial intelligence; Benchmarking; Learning systems; Statistics; Approximate inference; Expectation; Propagation; Feature selection methods; Latent variable; Learning tasks; Model-based OPC; Predictive performance; Probabilistic modeling; Feature extraction; InformáticaRights
Copyright 2015 by the author(s)Abstract
Multi-task feature selection methods often make
the hypothesis that learning tasks share relevant
and irrelevant features. However, this hypothesis
may be too restrictive in practice. For example,
there may be a few tasks with specific relevant
and irrelevant features (outlier tasks). Similarly,
a few of the features may be relevant for
only some of the tasks (outlier features). To account
for this, we propose a model for multi-task
feature selection based on a robust prior distribution
that introduces a set of binary latent variables
to identify outlier tasks and outlier features.
Expectation propagation can be used for efficient
approximate inference under the proposed prior.
Several experiments show that a model based on
the new robust prior provides better predictive
performance than other benchmark methods.
Files in this item
Google Scholar:Hernández Lobato, Daniel
-
Hernández Lobato, José Miguel
-
Ghahramani, Zoubin
This item appears in the following Collection(s)
Related items
Showing items related by title, author, creator and subject.