Searching for Quality Diversity When Diversity is Unaligned with Quality

Inspired by natural evolution’s affinity for discovering a wide variety of successful organisms, a new evolutionary search paradigm has emerged wherein the goal is not to find the single best solution but rather to collect a diversity of unique phenotypes where each variant is as good as it can be. These quality diversity (QD) algorithms therefore must explore multiple promising niches simultaneously. A QD algorithm’s diversity component, formalized by specifying a behavior characterization (BC), not only generates diversity but also promotes quality by helping to overcome deception in the fitness landscape. However, some BCs (particularly those that are unaligned with the notion of quality) do not adequately mitigate deception, rendering QD algorithms unable to discover the best-performing solutions on difficult problems. This paper introduces a solution that enables QD algorithms to pursue arbitrary notions of diversity without compromising their ability to solve hard problems: driving search with multiple BCs simultaneously.

[1]  Julian Togelius,et al.  Sentient Sketchbook: Computer-aided game level authoring , 2013, FDG.

[2]  Antoine Cully,et al.  Robots that can adapt like animals , 2014, Nature.

[3]  Jean-Baptiste Mouret,et al.  Illuminating search spaces by mapping elites , 2015, ArXiv.

[4]  Anders Lyhne Christensen,et al.  Evolution of swarm robotics systems with novelty search , 2013, Swarm Intelligence.

[5]  Anders Lyhne Christensen,et al.  Devising Effective Novelty Search Algorithms: A Comprehensive Empirical Study , 2015, GECCO.

[6]  Kenneth O. Stanley,et al.  Confronting the Challenge of Quality Diversity , 2015, GECCO.

[7]  Jason Yosinski,et al.  Deep neural networks are easily fooled: High confidence predictions for unrecognizable images , 2014, 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[8]  Shimon Whiteson,et al.  Critical factors in the performance of novelty search , 2011, GECCO '11.

[9]  Kenneth O. Stanley,et al.  Evolving a diversity of virtual creatures through novelty search and local competition , 2011, GECCO '11.

[10]  Kenneth O. Stanley,et al.  Abandoning Objectives: Evolution Through the Search for Novelty Alone , 2011, Evolutionary Computation.

[11]  Risto Miikkulainen,et al.  Evolving Neural Networks through Augmenting Topologies , 2002, Evolutionary Computation.

[12]  Kalyanmoy Deb,et al.  A fast and elitist multiobjective genetic algorithm: NSGA-II , 2002, IEEE Trans. Evol. Comput..

[13]  Jason Yosinski,et al.  Innovation Engines: Automated Creativity and Improved Stochastic Optimization via Deep Learning , 2015, GECCO.

[14]  Sebastian Risi,et al.  Single-unit pattern generators for quadruped locomotion , 2013, GECCO '13.

[15]  Kenneth O. Stanley,et al.  Indirectly Encoded Sodarace for Artificial Life , 2013, ECAL.

[16]  Stéphane Doncieux,et al.  Encouraging Behavioral Diversity in Evolutionary Robotics: An Empirical Study , 2012, Evolutionary Computation.

[17]  Kenneth O. Stanley,et al.  Unsupervised Feature Learning through Divergent Discriminative Feature Accumulation , 2014, AAAI.