Lutton, Evelyne, and Jean-Daniel Fekete. “Visual analytics and experimental analysis of evolutionary algorithms.” PhD diss., INRIA, 2011.
URL1 URL2
Experimental analysis of evolutionary algorithms usually aims at tuning the parameter setting or at improving knowledge about internal mechanisms (operators efficiency, genetic material distribution, or diversity management for instance). This crucial step relies on the analysis of a huge amount of multidimensional data, including numeric and symbolic data. Usual features of existing EA visualisation systems consist in visualising time- or generation-dependent curves (fitness, diversity, or other statistics). But when dealing with detailed genomic information, the task becomes more difficult, as a convenient visualisation strongly depends on the considered fitness landscape. In this latter case the raw data are usually sets of successive populations of points of a complex multidimensional space. The purpose of this paper is to evaluate the potential interest of some recent visual analytics tools for navigating in complex sets of EA data, and to sketch future developements of visual analytics tools adapted to the needs of EA experimental analysis.