3-5 Jul 2013 Villeneuve d'Ascq (Lille) (France)

By author > Cornuéjols Antoine

Thursday 4
Machine Learning

› 16:30 - 17:00 (30min)
Anticipative and Dynamic Adaptation to Concept Changes
Antoine Cornuéjols  1, *@  , Ghazal Jaber  1, *@  , Philippe Tarroux  2@  
1 : AgroParisTech, UMR 518 MIA  -  Website
AgroParisTech
AgroParisTech, dept. MMIP 16, rue Claude Bernard F-75231 Paris Cedex 05 -  France
2 : Laboratoire d'Informatique pour la Mécanique et les Sciences de l'Ingénieur [Orsay]  (LIMSI)  -  Website
Université Pierre et Marie Curie (UPMC) - Paris VI, Université Paris XI - Paris Sud, CNRS : UPR3251, Université Pierre et Marie Curie [UPMC] - Paris VI
Université Paris Sud (Paris XI) Bât. 508 BP 133 91403 ORSAY CEDEX -  France
* : Corresponding author

Learning from data streams is emerging as an important application area. When the environment changes, as is increasingly the case when considering unending streams and long-life learning, it is necessary to rely on on-line learning with the capability to adapt to changing conditions a.k.a. concept drifts. Previous works have focused on means to detect changes and to adapt to them. Ensemble methods relying on committees of base learners have been among the most successful approaches.

In this paper, we go one step further by introducing a second-order learning mechanism that is able to detect relevant states of the environment, to recognize recurring contexts and to anticipate likely concepts changes. Results of an empirical comparison with adaptive methods show that, for a very slight price in memory and computation load, the proposed algorithm always improves on, or at worst equals, the prediction performance of a mere adaptive approach.


Online user: 1