Data-Efficient Design Exploration through Surrogate-Assisted Illumination

• Published in: Evolutionary computation Vol. 26; no. 3; pp. 381 - 410
• Main Authors: Gaier, Adam, Asteroth, Alexander, Mouret, Jean-Baptiste
• Format: Journal Article
• Language: English
• Published: One Rogers Street, Cambridge, MA 02142-1209, USA MIT Press 01.09.2018; MIT Press Journals, The; Massachusetts Institute of Technology Press (MIT Press)
• Subjects: Algorithms; Artificial Intelligence; Bayes Theorem; Biomechanical Phenomena; Computational Engineering, Finance, and Science; Computer Aided Engineering; computer automated design; Computer Science; Computer Simulation; Data Interpretation, Statistical; Design optimization; Domains; Engineering Sciences; Exploration; Extended Papers Invited from GECCO 2017 (Berlin); Flow simulation; Fluids mechanics; Illumination; Lighting; Machine Learning; MAP-Elites; Mechanics; Optimization techniques; Quality Control; quality-diversity; Robotics; Software; Statistics; surrogate modeling; quality-diversity; surrogate modeling; computer automated design; MAP-Elites; Computer Automated Design; Surrogate Modeling; Quality Diversity
• ISSN: 1063-6560; 1530-9304; 1530-9304
• DOI: 10.1162/evco_a_00231

Design optimization techniques are often used at the beginning of the design process to explore the space of possible designs. In these domains illumination algorithms, such as MAP-Elites, are promising alternatives to classic optimization algorithms because they produce diverse, high-quality solutions in a single run, instead of only a single near-optimal solution. Unfortunately, these algorithms currently require a large number of function evaluations, limiting their applicability. In this article, we introduce a new illumination algorithm, Surrogate-Assisted Illumination (SAIL), that leverages surrogate modeling techniques to create a map of the design space according to user-defined features while minimizing the number of fitness evaluations. On a two-dimensional airfoil optimization problem, SAIL produces hundreds of diverse but high-performing designs with several orders of magnitude fewer evaluations than MAP-Elites or CMA-ES. We demonstrate that SAIL is also capable of producing maps of high-performing designs in realistic three-dimensional aerodynamic tasks with an accurate flow simulation. Data-efficient design exploration with SAIL can help designers understand what is possible, beyond what is optimal, by considering more than pure objective-based optimization.