EMACI: A Visual Interface of Multi-Parameters for Interactive Evolutionary Algorithm

This paper proposes a technique for visualizing the progression of evolutionary algorithms such as genetic algorithms (GAs). Our technique supposes that a GA solves optimization problems that maximize target functions with multiple parameters. Then, our technique presents the timevarying parameters during the progression of the GA. The technique applies two-tone pseudo-coloring for the precise representation of changes in parameter values, so that users can understand the behavior of the GA. We also provide a user interface to control the configuration of the GA so that users can obtain the optimal solution earlier. The paper shows experimental results to demonstrate the efficacy of the technique.