PROJECT TITLE :

Robust Optimization Over Time: Problem Difficulties and Benchmark Problems

ABSTRACT:

The focus of most research in evolutionary dynamic optimization has been tracking moving optimum (TMO). However, TMO will not capture all the characteristics of real-world dynamic optimization problems (DOPs), especially in situations where a resolution's future fitness needs to be thought-about. To account for a resolution's future fitness explicitly, we propose to seek out sturdy solutions to DOPs, which are formulated as the robust optimization over time (ROOT) downside. During this paper we have a tendency to analyze 2 robustness definitions in ROOT and then develop two varieties of benchmark issues for the two robustness definitions in ROOT, respectively. The 2 types of benchmark problems are motivated by the inappropriateness of existing DOP benchmarks for the study of ROOT. Additionally, we evaluate four representative strategies from the literature on our proposed ROOT benchmarks, so as to realize a better understanding of ROOT issues and their relationship to additional in style TMO problems. The experimental results are analyzed, that show the strengths and weaknesses of various methods in solving ROOT problems with totally different dynamics. In particular, the important challenges of ROOT problems have been revealed for the first time by the experimental results on our proposed ROOT benchmarks.