Sciweavers

Share
ICML
2010
IEEE

Feature Selection Using Regularization in Approximate Linear Programs for Markov Decision Processes

8 years 10 months ago
Feature Selection Using Regularization in Approximate Linear Programs for Markov Decision Processes
Approximate dynamic programming has been used successfully in a large variety of domains, but it relies on a small set of provided approximation features to calculate solutions reliably. Large and rich sets of features can cause existing algorithms to overfit because of a limited number of samples. We address this shortcoming using L1 regularization in approximate linear programming. Because the proposed method can automatically select the appropriate richness of features, its performance does not degrade with an increasing number of features. These results rely on new and stronger sampling bounds for regularized approximate linear programs. We also propose a computationally efficient homotopy method. The empirical evaluation of the approach shows that the proposed method performs well on simple MDPs and standard benchmark problems.
Marek Petrik, Gavin Taylor, Ronald Parr, Shlomo Zi
Added 09 Nov 2010
Updated 09 Nov 2010
Type Conference
Year 2010
Where ICML
Authors Marek Petrik, Gavin Taylor, Ronald Parr, Shlomo Zilberstein
Comments (0)
books