CMU-ML-07-118
Machine Learning Department
School of Computer Science, Carnegie Mellon University



CMU-ML-07-118

Selecting Observations against Adversarial Objectives

Andreas Krause, H. Brendan McMahan*
Carlos Guestrin, Anupam Gupta

December 2007

CMU-ML-07-118.pdf


Keywords: Gaussian processes, experimental design, active learning, submodular functions, observation selection


In many applications, one has to actively select among a set of expensive observations before making an informed decision. Often, we want to select observations which perform well when evaluated with an objective function chosen by an adversary. Examples include minimizing the maximum posterior variance in Gaussian Process regression, robust experimental design, and sensor placement for outbreak detection. In this paper, we present the Submodular Saturation algorithm, a simple and efficient algorithm with strong theoretical approximation guarantees for the case where the possible objective functions exhibit submodularity, an intuitive diminishing returns property. Moreover,we prove that better approximation algorithms do not exist unless NP-complete problems admit efficient algorithms. We evaluate our algorithm on several real-world problems. For Gaussian Process regression, our algorithm compares favorably with state-of-the-art heuristics described in the geostatistics literature, while being simpler, faster and providing theoretical guarantees. For robust experimental design, our algorithm performs favorably compared to SDP-based algorithms.

16 pages

*Google Pittsburgh


SCS Technical Report Collection
School of Computer Science homepage

This page maintained by reports@cs.cmu.edu