Knowing What to Ask: A Bayesian Active Learning Approach to the Surveying Problem

Authors

  • Yoad Lewenberg The Hebrew University of Jerusalem
  • Yoram Bachrach Digital Genius Ltd.
  • Ulrich Paquet Microsoft Research, Cambridge
  • Jeffrey Rosenschein The Hebrew University of Jerusalem

DOI:

https://doi.org/10.1609/aaai.v31i1.10730

Keywords:

Probabilistic Graphical Models, Active Surveying

Abstract

We examine the surveying problem, where we attempt to predict how a target user is likely to respond to questions by iteratively querying that user, collaboratively based on the responses of a sample set of users. We focus on an active learning approach, where the next question we select to ask the user depends on their responses to the previous questions. We propose a method for solving the problem based on a Bayesian dimensionality reduction technique. We empirically evaluate our method, contrasting it to benchmark approaches based on augmented linear regression, and show that it achieves much better predictive performance, and is much more robust when there is missing data.

Downloads

Published

2017-02-12

How to Cite

Lewenberg, Y., Bachrach, Y., Paquet, U., & Rosenschein, J. (2017). Knowing What to Ask: A Bayesian Active Learning Approach to the Surveying Problem. Proceedings of the AAAI Conference on Artificial Intelligence, 31(1). https://doi.org/10.1609/aaai.v31i1.10730

Issue

Section

Main Track: Machine Learning Applications