Sequential approaches for learning datum-wise sparse representations

Abstract : In supervised classification, data representation is usually considered at the dataset level: one looks for the "best" representation of data assuming it to be the same for all the data in the data space. We propose a different approach where the representations used for classification are tailored to each datum in the data space. One immediate goal is to obtain sparse datum-wise representations: our approach learns to build a representation specific to each datum that contains only a small subset of the features, thus allowing classification to be fast and efficient. This representation is obtained by way of a sequential decision process that sequentially chooses which features to acquire before classifying a particular point; this process is learned through algorithms based on Reinforcement Learning. The proposed method performs well on an ensemble of medium-sized sparse classification problems. It offers an alternative to global sparsity approaches, and is a natural framework for sequential classification problems. The method extends easily to a whole family of sparsity-related problem which would otherwise require developing specific solutions. This is the case in particular for cost-sensitive and limited-budget classification, where feature acquisition is costly and is often performed sequentially. Finally, our approach can handle non-differentiable loss functions or combinatorial optimization encountered in more complex feature selection problems.
Document type :
Journal articles
Machine Learning, Springer Verlag (Germany), 2012, 89 (1-2), pp.87-122. <http://link.springer.com/article/10.1007%2Fs10994-012-5306-7>. <10.1007/s10994-012-5306-7>


https://hal.inria.fr/hal-00747724
Contributor : Preux Philippe <>
Submitted on : Thursday, November 8, 2012 - 3:25:43 PM
Last modification on : Thursday, November 8, 2012 - 3:39:43 PM

File

versionPublieeMLJ.pdf
fileSource_public_greenPublisher

Identifiers

Collections

Citation

Gabriel Dulac-Arnold, Ludovic Denoyer, Philippe Preux, Patrick Gallinari. Sequential approaches for learning datum-wise sparse representations. Machine Learning, Springer Verlag (Germany), 2012, 89 (1-2), pp.87-122. <http://link.springer.com/article/10.1007%2Fs10994-012-5306-7>. <10.1007/s10994-012-5306-7>. <hal-00747724>

Export

Share

Metrics

Consultation de
la notice

270

Téléchargement du document

78