Near-optimal Nonmyopic Value of Information in Graphical Models
arXiv:1207.1394
Abstract
A fundamental issue in real-world systems, such as sensor networks, is the selection of observations which most effectively reduce uncertainty. More specifically, we address the long standing problem of nonmyopically selecting the most informative subset of variables in a graphical model. We present the first efficient randomized algorithm providing a constant factor (1-1/e-epsilon) approximation guarantee for any epsilon > 0 with high confidence. The algorithm leverages the theory of submodular functions, in combination with a polynomial bound on sample complexity. We furthermore prove that no polynomial time algorithm can provide a constant factor approximation better than (1 - 1/e) unless P = NP. Finally, we provide extensive evidence of the effectiveness of our method on two complex real-world datasets.
Appears in Proceedings of the Twenty-First Conference on Uncertainty in Artificial Intelligence (UAI2005)
References in corpus (1)
Cited by in corpus (13)
- Learning Mixtures of Submodular Shells with Application to Document Summarization
- Parallelizing Exploration-Exploitation Tradeoffs with Gaussian Process Bandit Optimization
- Structured sparsity-inducing norms through submodular functions
- Expectation-Maximization for Learning Determinantal Point Processes
- Convex Analysis and Optimization with Submodular Functions: a Tutorial
- Constrained Submodular Maximization via a Non-symmetric Technique
- Convex Relaxations for Learning Bounded Treewidth Decomposable Graphs
- Active Classification: Theory and Application to Underwater Inspection
- Scalable Variational Inference in Log-supermodular Models
- Observation Subset Selection as Local Compilation of Performance Profiles
- Active Learning for Autonomous Intelligent Agents: Exploration, Curiosity, and Interaction
- Practical Context Awareness: Measuring and Utilizing the Context Dependency of Mobile Usage
- Online Distributed Sensor Selection