Budgeted Learning of Bounded Active Classifiers
- Aloak Kapoor, Dept of Computing Science, University of Alberta
- Russ Greiner, Dept of Computing Science; PI of AICML
Since resources for data acquisition are seldom infinite, both learners and classifiers must act intelligently under hard budgets. In this paper, we consider problems in which feature values are unknown to both the learner and classifier, but can be acquired at a cost. Our goal is a learner that spends its fixed learning budget bL acquiring training data, to produce the most accurate ``active classifier'' that spends at most bC per instance. To produce this fixed-budget classifier, the fixed-budget learner needs to sequentially decide which feature values to collect in order to learn the relevant information about the underlying distribution. We explore a variety of approaches the learner can take, including the standard ``round robin'' policy (purchasing every feature of every instance until the bL budget is exhausted). In this work, we demonstrate empirically that the round robin strategy is problematic (especially for small bL), and provide alternate learning strategies that achieve superior performance on a variety of real-world datasets.
Citation
A. Kapoor, R. Greiner. "Budgeted Learning of Bounded Active Classifiers". Utility-Based Data Mining (UBDM), August 2005.| Keywords: | active classifier, budgeted learning, machine learning |
| Category: | In Workshop |
BibTeX
@misc{Kapoor+Greiner:UBDM05,
author = {Aloak Kapoor and Russ Greiner},
title = {Budgeted Learning of Bounded Active Classifiers},
booktitle = {Utility-Based Data Mining (UBDM)},
year = 2005,
}Last Updated: August 13, 2007Submitted by Russ Greiner
