Information Acquisition and Utilization Problems

This paper focuses on the problem of information acquisition and utilization where a decision maker, by carefully controlling a sequence of actions with uncertain outcomes, dynamically refines his/her belief about stochastically time-varying parameters of interest in order to utilize a system of interest as efficiently as possible. The paper proposes a new theoretical framework for stochastic learning and decision-making in such a setting termed Information Acquisition and Utilization Problems (IAUP). IAUP is a special case of partially observable Markov decision problems (POMDP) with several unique properties. Motivated by a synthesis of the prior works on active hypothesis testing, noisy dynamic search, and joint source-channel coding, this framework borrows from diverse areas of research from statistics, information theory, and stochastic control. The main contribution of this work is to establish the general attributes of IAUPs as well as their connections to known fields of study.