A Collaborative 20 Questions Model for Target Search with Human-Machine Interaction
Abstract
We consider the problem of 20 questions with noise for collaborative players under the minimum entropy criterion in the setting of stochastic search, with application to target localization. First, assuming conditionally independent collaborators, we characterize the structure of the optimal policy for constructing the sequence of questions. This generalizes the single player probabilistic bisection method for stochastic search problems. Second, we prove a separation theorem showing that optimal joint queries achieve the same performance as a greedy sequential scheme. Third, we establish convergence rates of the mean-squared error (MSE). Fourth, we derive upper bounds on the MSE of the sequential scheme. This framework provides a mathematical model for incorporating a human in the loop for active machine learning systems.
Document Details
- Document Type
- Technical Report
- Publication Date
- May 01, 2013
- Accession Number
- ADA581713
Entities
People
- Alfred O. Hero III
- Brian M. Sadler
- Theodoros Tsiligkaridis
Organizations
- University of Michigan