Boosting Active Learning to Optimality: a Tractable Monte-Carlo, Billiard-based Algorithm

Philippe Rolet 1 Michèle Sebag 2 Olivier Teytaud 1, 2, 3
2 TAO - Machine Learning and Optimisation
LRI - Laboratoire de Recherche en Informatique, UP11 - Université Paris-Sud - Paris 11, Inria Saclay - Ile de France, CNRS - Centre National de la Recherche Scientifique : UMR8623
3 TANC - Algorithmic number theory for cryptology
Inria Saclay - Ile de France, LIX - Laboratoire d'informatique de l'École polytechnique [Palaiseau]
Abstract : Abstract. This paper focuses on Active Learning with a limited num- ber of queries; in application domains such as Numerical Engineering, the size of the training set might be limited to a few dozen or hundred exam- ples due to computational constraints. Active Learning under bounded resources is formalized as a finite horizon Reinforcement Learning prob- lem, where the sampling strategy aims at minimizing the expectation of the generalization error. A tractable approximation of the optimal (in- tractable) policy is presented, the Bandit-based Active Learner (BAAL) algorithm. Viewing Active Learning as a single-player game, BAAL com- bines UCT, the tree structured multi-armed bandit algorithm proposed by Kocsis and Szepesv´ri (2006), and billiard algorithms. A proof of a principle of the approach demonstrates its good empirical convergence toward an optimal policy and its ability to incorporate prior AL crite- ria. Its hybridization with the Query-by-Committee approach is found to improve on both stand-alone BAAL and stand-alone QbC.
Document type :
Conference papers
Complete list of metadatas

Cited literature [36 references]  Display  Hide  Download

https://hal.inria.fr/inria-00433866
Contributor : Olivier Teytaud <>
Submitted on : Friday, November 20, 2009 - 1:17:05 PM
Last modification on : Wednesday, March 27, 2019 - 4:41:29 PM
Long-term archiving on : Thursday, June 30, 2011 - 11:57:56 AM

File

BALO.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : inria-00433866, version 1

Collections

Citation

Philippe Rolet, Michèle Sebag, Olivier Teytaud. Boosting Active Learning to Optimality: a Tractable Monte-Carlo, Billiard-based Algorithm. ECML, 2009, Bled, Slovenia. pp.302-317. ⟨inria-00433866⟩

Share

Metrics

Record views

2570

Files downloads

860