EvoSpace-i: a framework for interactive evolutionary algorithms

  • Authors:
  • Mario García-Valdez;Juan J. Merelo;Leonardo Trujillo;Francisco Fernández-de-Vega;José C. Romero;Alejandra Mancilla

  • Affiliations:
  • Instituto Tecnológico de Tijuana, Tijuana, México;University of Granada, Granada, Spain;Instituto Tecnológico de Tijuana, Tijuana, México;Universidad de Extremadura, Extremadura, Spain;Instituto Tecnológico de Tijuana, Tijuana, México;Instituto Tecnológico de Tijuana, Tijuana, México

  • Venue:
  • Proceedings of the 15th annual conference companion on Genetic and evolutionary computation
  • Year:
  • 2013

Quantified Score

Hi-index 0.00

Visualization

Abstract

Evolutionary art (EvoArt) encompasses a variety of research devoted to the development of evolutionary systems that can help produce artistic artifacts in an automated or semi-automated process. Given the difficulty of evaluating subjective artistic preferences, one of the main approaches used by EvoArt researchers is interactive evolution where user input guides the search. However, despite the growth of EvoArt over recent years the research area still lacks a comprehensive software tool that can help in the development of EvoArt applications. Therefore, this work presents EvoSpace-i, an open source framework for the development of collaborative-interactive evolutionary algorithms for art and design. The main components of the framework are: (i) Evospace, a population store for the development of cloud-based evolutionary algorithms, implemented using Re-dis key-value server; and an (ii) Interactive web application where end-users collaborate in a social network sharing, collecting, rating and ultimately evolving individuals. Individuals can be presented as multimedia elements or artistic artifacts (images, animations, sound) using the Processing programming language, a development language specifically aimed at artists. EvoSpace-i is designed to be easy to use and setup, allowing researchers, and more importantly artists, to quickly develop distributed and collaborative EvoArt applications. This paper presents the main details of EvoSpace-i and two example applications to illustrate the potential of the tool.