Feature Selection as a One-Player Game - Archive ouverte HAL Access content directly
Conference Papers Year : 2010

Feature Selection as a One-Player Game

(1) , (1, 2)


This paper formalizes Feature Selection as a Reinforcement Learning problem, leading to a provably optimal though intractable selection policy. As a second contribution, this paper presents an approximation thereof, based on a one-player game approach and relying on the Monte-Carlo tree search UCT (Upper Confidence Tree) proposed by Kocsis and Szepesvari (2006). The Feature Uct SElection (FUSE) algorithm extends UCT to deal with i) a finite unknown horizon (the target number of relevant features); ii) the huge branching factor of the search tree, reflecting the size of the feature set. Finally, a frugal reward function is proposed as a rough but unbiased estimate of the relevance of a feature subset. A proof of concept of FUSE is shown on benchmark data sets.
Fichier principal
Vignette du fichier
fuse_icml10.pdf (226.85 Ko) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

inria-00484049 , version 1 (17-05-2010)


  • HAL Id : inria-00484049 , version 1


Romaric Gaudel, Michèle Sebag. Feature Selection as a One-Player Game. International Conference on Machine Learning, Jun 2010, Haifa, Israel. pp.359--366. ⟨inria-00484049⟩
4313 View
3452 Download


Gmail Facebook Twitter LinkedIn More