A Flexible Approach to Quantifying Various Dimensions of Environmental Complexity
Michael L. Anderson · 2004
ABSTRACT — In this paper I propose a flexible method of quantifying various dimensions of the complexity of a test environment, including its information density, variability, volatility, inconsistency, and uncertainty. This allows one to determine the task performance of intelligent agents as a function of such measures, and therefore permits derivative measures of their perturbation tolerance—that is, their ability to cope with a complex and changing environment. Keywords: complexity, perturbation tolerance, standards, testbeds 1.