Cost-Bounded Active Classification Using Partially Observable Markov Decision Processes

TL;DR

This paper develops a decision-theoretic framework using POMDPs for active classification of dynamical systems modeled as MDPs, aiming for efficient and confident model identification within cost and time constraints.

Contribution

It introduces a novel POMDP-based approach for active classification of MDP models, including exact and approximate strategies for decision-making under cost and confidence constraints.

Findings

Exact strategy computed via value iteration

Approximate strategy using adaptive sampling

Successful application in medical diagnosis and intrusion detection

Abstract

Active classification, i.e., the sequential decision-making process aimed at data acquisition for classification purposes, arises naturally in many applications, including medical diagnosis, intrusion detection, and object tracking. In this work, we study the problem of actively classifying dynamical systems with a finite set of Markov decision process (MDP) models. We are interested in finding strategies that actively interact with the dynamical system, and observe its reactions so that the true model is determined efficiently with high confidence. To this end, we present a decision-theoretic framework based on partially observable Markov decision processes (POMDPs). The proposed framework relies on assigning a classification belief (a probability distribution) to each candidate MDP model. Given an initial belief, some misclassification probabilities, a cost bound, and a finite time horizon, we design POMDP strategies leading to classification decisions. We present two different approaches to find such strategies. The first approach computes the optimal strategy "exactly" using value iteration. To overcome the computational complexity of finding exact solutions, the second approach is based on adaptive sampling to approximate the optimal probability of reaching a classification decision. We illustrate the proposed methodology using two examples from medical diagnosis and intruder detection.