Action Selection Under Uncertainty and Risk Involving High Consequence Rare Events

Abstract

This paper describes an approach to decision making for action selection in the presence of high consequence rare events. A model presented in the paper is an adaptation of the Multiple Quantile Theory that evaluates actions based on Choquet utility while considering decision making attitudes towards gains and losses. In our model, the capacities used for computing Choquet utility are based on the belief structure resulting from situation assessment, which allows for eliminating a priori parameter selection required by the Multiple Quantile Theory. A use case designed to illustrate the feasibility of the model in a simple military scenario is also presented.