Learning Diagnostic Policies from Examples by Systematic Search

TL;DR

This paper formalizes the process of learning cost-sensitive diagnostic policies as a Markov Decision Process and compares systematic search and greedy algorithms, demonstrating the superiority of systematic search on benchmark datasets.

Contribution

It introduces a novel integration of overfitting prevention into systematic search for diagnostic policies and compares two algorithms for solving the MDP.

Findings

Systematic search outperforms greedy methods in policy quality.

Regularizers effectively prevent overfitting and accelerate search.

The approach is validated on benchmark datasets.

Abstract

A diagnostic policy specifies what test to perform next, based on the results of previous tests, and when to stop and make a diagnosis. Cost-sensitive diagnostic policies perform tradeoffs between (a) the cost of tests and (b) the cost of misdiagnoses. An optimal diagnostic policy minimizes the expected total cost. We formalize this diagnosis process as a Markov Decision Process (MDP). We investigate two types of algorithms for solving this MDP: systematic search based on AO* algorithm and greedy search (particularly the Value of Information method). We investigate the issue of learning the MDP probabilities from examples, but only as they are relevant to the search for good policies. We do not learn nor assume a Bayesian network for the diagnosis process. Regularizers are developed to control overfitting and speed up the search. This research is the first that integrates overfitting prevention into systematic search. The paper has two contributions: it discusses the factors that make systematic search feasible for diagnosis, and it shows experimentally, on benchmark data sets, that systematic search methods produce better diagnostic policies than greedy methods.