Multi-Objective Optimization for Hard Real-World Problems under Limited Budgets

Sequential parameter optimization (SPO) is a heuristic that combines classical and modern statistical techniques to improve the performance of search algorithms and optimization tasks. Since only a fraction of the interesting industrial Problems are single objective it is fundamental to provide algorithms with the SPO-Toolbox (SPOT) to optimize multi objective problems. Moreover, with the resulting capability to allow competing objectives, the opportunity arises to not only aim for the best, but also for the most robust solution. In this Paper we show that multi objective optimization can be e ciently done with SPOT. Furthermore we present an approach to optimize not only the quality of the solution, but also its robustness, taken these two goals as criteria for multi objective optimization. The results that we are presenting show...