Testing Monte Carlo Algorithmic Systems

This article covers the unique challenge in defining testing scope and expected results when testing systems with non-deterministic outputs whose accuracy improves over repeated iterations of the same inputs. A thorough understanding of the algorithms under test and excellent communication between development and testing are essential in test scenario definition and predicting anticipated outcomes. Defining tests and expected behaviors prior to the start of testing is especially crucial in these types of conditions.