Software management oracles often contain numerous subjective features. At each subjective point, a range of behaviors is possible. Stochastic simulation samples a subset of the possible behaviors. After many such stochastic simulations, the TAR2 treatment learner can find control actions that have (usually) the same impact despite the subjectivity of the oracle.
No takes yet. Share an insight, caveat, or question.
Menzies et al. (2005) studied this question.