On the performance assessment and comparison of stochastic multiobjective optimizers

This work proposes a quantitative, non-parametric interpretation of statistical performance of stochastic multiobjective optimizers, including, but not limited to, genetic algorithms. It is shown that, according to this interpretation, typical performance can be defined in terms analogous to the notion of median for ordinal data, as can other measures analogous to other quantiles.Non-parametric statistical test procedures are then shown to be useful in deciding the relative performance of different multiobjective optimizers on a given problem. Illustrative experimental results are provided to support the discussion.