Skip to Main content Skip to Navigation
Conference papers

Active learning in regression, with an application to stochastic dynamic programming

Olivier Teytaud 1 Sylvain Gelly 1 Jérémie Mary 1
1 TANC - Algorithmic number theory for cryptology
Inria Saclay - Ile de France, LIX - Laboratoire d'informatique de l'École polytechnique [Palaiseau]
Abstract : We study active learning as a derandomized form of sampling. We show that full derandomization is not suitable in a robust framework, propose partially derandomized samplings, and develop new active learning methods (i) in which expert knowledge is easy to integrate (ii) with a parameter for the exploration/exploitation dilemma (iii) less randomized than the full-random sampling (yet also not deterministic). Experiments are performed in the case of regression for value-function learning on a continuous domain. Our main results are (i) efficient partially derandomized point sets (ii) moderate-derandomization theorems (iii) experimental evidence of the importance of the frontier (iv) a new regression-specific user-friendly sampling tool lessrobust than blind samplers but that sometimes works very efficiently in large dimensions. All experiments can be reproduced by downloading the source code and running the provided command line.
Document type :
Conference papers
Complete list of metadata

Cited literature [30 references]  Display  Hide  Download

https://hal.inria.fr/inria-00173204
Contributor : Olivier Teytaud <>
Submitted on : Wednesday, September 19, 2007 - 2:15:17 PM
Last modification on : Thursday, December 3, 2020 - 4:56:03 PM
Long-term archiving on: : Friday, April 9, 2010 - 2:28:23 AM

File

ldsfordp.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : inria-00173204, version 1

Collections

Citation

Olivier Teytaud, Sylvain Gelly, Jérémie Mary. Active learning in regression, with an application to stochastic dynamic programming. ICINCO 2007, 2007, Angers, France. ⟨inria-00173204⟩

Share

Metrics

Record views

415

Files downloads

655