| Thème(s) : |
 |
|
|
| Titre : |
|
Fast classification using sparse decision DAGs |
|
| Auteur : |
|
R. Busa-Fekete1, 2, D. Benbouzid1, 2, B. Kégl1, 2, 3 |
|
| Laboratoire : |
|
|
|
| équipe(s) de recherche : |
|
APPSTAT |
| Résumé : |
|
In this paper we propose an algorithm that builds sparse decision DAGs (directed acyclic graphs) out of a list of base classifiers provided by an external learning method such as AdaBoost. The basic idea is to cast the DAG design task as a Markov decision process. Each instance can decide to use or to skip each base classifier, based on the current state of the classifier being built. The result is a sparse decision DAG where the base classifiers are selected in a data-dependent way. The method has a single hyperparameter with a clear semantics of controlling the accuracy/speed trade-off. The algorithm is competitive with state-of-the-art cascade detectors on three object-detection benchmarks, and it clearly outperforms them in the regime of low number of base classifiers. Unlike cascades, it is also readily applicable for multi-class classification. Using the multi-class setup, we show on a benchmark web page ranking data set that we can significantly improve the decision speed without harming the performance of the ranker. |
|
|
| Type de publication : |
|
Communications sans actes |
|
| Date de la présentation : |
|
29/06/2012 |
|
|
| Nom du colloque : |
|
29th International Conference on Machine Learning (ICML 2012) |
|
| Ville du colloque : |
|
Edinburgh |
|
| Pays du colloque : |
|
Royaume-Uni |
|
| Date du colloque (début) : |
|
26/06/2012 |
|
| Date du colloque (fin) : |
|
01/07/2012 |
|
|
| Identifiant local : |
|
LAL 12-223 |
|
|
Récupérer au format
