Active learning via transductive experimental design

14 years 7 months ago
Active learning via transductive experimental design
This paper considers the problem of selecting the most informative experiments x to get measurements y for learning a regression model y = f(x). We propose a novel and simple concept for active learning, transductive experimental design, that explores available unmeasured experiments (i.e.,unlabeled data) and has a better scalability in comparison with classic experimental design methods. Our in-depth analysis shows that the new method tends to favor experiments that are on the one side hardto-predict and on the other side representative for the rest of the experiments. Efficient optimization of the new design problem is achieved through alternating optimization and sequential greedy search. Extensive experimental results on synthetic problems and three real-world tasks, including questionnaire design for preference learning, active learning for text categorization, and spatial sensor placement, highlight the advantages of the proposed approaches.
Kai Yu, Jinbo Bi, Volker Tresp
Added 17 Nov 2009
Updated 17 Nov 2009
Type Conference
Year 2006
Where ICML
Authors Kai Yu, Jinbo Bi, Volker Tresp
Comments (0)