Active Exploration via Experiment Design in Markov Chains

TL;DR

This paper introduces extsc{markov-design}, an algorithm for optimal experiment design in Markov chain settings, enabling adaptive policy selection with proven convergence to the best measurement allocation.

Contribution

We develop a sequential, adaptive algorithm for experiment design in Markov chains that guarantees convergence to the optimal policy, applicable to various real-world domains.

Findings

Algorithm provably converges to optimal measurement policy

Demonstrated effectiveness in ecological surveillance

Validated in pharmacology applications

Abstract

A key challenge in science and engineering is to design experiments to learn about some unknown quantity of interest. Classical experimental design optimally allocates the experimental budget to maximize a notion of utility (e.g., reduction in uncertainty about the unknown quantity). We consider a rich setting, where the experiments are associated with states in a {\em Markov chain}, and we can only choose them by selecting a {\em policy} controlling the state transitions. This problem captures important applications, from exploration in reinforcement learning to spatial monitoring tasks. We propose an algorithm -- \textsc{markov-design} -- that efficiently selects policies whose measurement allocation \emph{provably converges to the optimal one}. The algorithm is sequential in nature, adapting its choice of policies (experiments) informed by past measurements. In addition to our theoretical analysis, we showcase our framework on applications in ecological surveillance and pharmacology.