Search Trajectories Networks of Multiobjective Evolutionary Algorithms

Understanding the search dynamics of multiobjective evolutionary algorithms (MOEAs) is still an open problem. This paper extends a recent network-based tool, trajectory networks (STNs), to model behavior MOEAs. Our approach uses idea decomposition, where problem transformed into several single-objective problems. We show that STNs can be used and distinguish two popular algorithms, MOEA/D NSGA-II, using 10 continuous benchmark problems with 2 3 objectives. findings suggest we improve our understanding MOEAs for algorithm analysis.