Data-Efficient Design Exploration through Surrogate-Assisted Illumination

被引:45
作者
Gaier, Adam [1 ,2 ]
Asteroth, Alexander [2 ]
Mouret, Jean-Baptiste [1 ]
机构
[1] Univ Lorraine, CNRS, INRIA, LORIA, F-54000 Nancy, France
[2] Bonn Rhein Sieg Univ Appl Sci, D-53757 St Augustin, Germany
基金
欧洲研究理事会;
关键词
MAP-Elites; surrogate modeling; quality-diversity; computer automated design; ALGORITHMS; EVOLUTION;
D O I
10.1162/evco_a_00231
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
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.
引用
收藏
页码:381 / 410
页数:30
相关论文
共 50 条
[1]  
[Anonymous], P 6 WORLD C STRUCT M
[2]  
[Anonymous], 2015, 2015 IEEE C COMP VIS
[3]  
[Anonymous], 2010, A tutorial on Bayesian optimization of expensive cost functions
[4]  
[Anonymous], EVOLVABLE SYSTEMS BI
[5]  
[Anonymous], 2006, GAUSSIAN PROCESSES M, DOI DOI 10.1142/S0129065704001899
[6]  
[Anonymous], 150404909 ARXIV
[7]  
Bradner E., 2014, P S SIM ARCH URB DES
[8]  
Calandra R, 2014, IEEE INT CONF ROBOT, P1951, DOI 10.1109/ICRA.2014.6907117
[9]   Reset-free Trial-and-Error Learning for Robot Damage Recovery [J].
Chatzilygeroudis, Konstantinos ;
Vassiliades, Vassilis ;
Mouret, Jean-Baptiste .
ROBOTICS AND AUTONOMOUS SYSTEMS, 2018, 100 :236-250
[10]   Robots that can adapt like animals [J].
Cully, Antoine ;
Clune, Jeff ;
Tarapore, Danesh ;
Mouret, Jean-Baptiste .
NATURE, 2015, 521 (7553) :503-U476