Pareto Front Upconvert by Iterative Estimation Modeling and Solution Sampling.

For an efficient upconvert of the Pareto front resolution by utilizing a known candidate solution set, this paper proposed an algorithm that built the Pareto front and the Pareto set estimation models and repeated to sample a solution from them, evaluate it, and updated the estimation models with it. Conventional supervised multi-objective optimization algorithm (SMOA) built the Pareto front and the Pareto set estimation models with a known candidate solution set. SMOA sampled a set of well-distributed estimated points and evaluated them to upconvert the Pareto front resolution. However, depending on the distribution of the known candidate solutions, we could not expect the accuracy of the estimation models and the estimated points from them. The proposed method, the iterative SMOA (I-SMOA), gradually improved the accuracy of the estimation models through their iterative update with evaluated solutions. Experimental results on the DTLZ2 test problem showed that the proposed I-SMOA obtained solutions uniformly distributed more than the one by the conventional SMOA, and the proposed I-SMOA achieved higher robustness on the initially given candidate solutions.