Automatically Designing State-of-the-Art Multi- and Many-Objective Evolutionary Algorithms

Evolutionary Computation - Tập 28 Số 2 - Trang 195-226 - 2020
Leonardo C. T. Bezerra1, Manuel López-Ibáñez2, Thomas Stützle3
1Instituto Metrópole Digital (IMD), Universidade Federal do Rio Grande do Norte, Natal, RN, Brazil
2Alliance Manchester Business School, University of Manchester, UK
3IRIDIA, CoDE, Université Libre de Bruxelles, Belgium

Tóm tắt

A recent comparison of well-established multiobjective evolutionary algorithms (MOEAs) has helped better identify the current state-of-the-art by considering (i) parameter tuning through automatic configuration, (ii) a wide range of different setups, and (iii) various performance metrics. Here, we automatically devise MOEAs with verified state-of-the-art performance for multi- and many-objective continuous optimization. Our work is based on two main considerations. The first is that high-performing algorithms can be obtained from a configurable algorithmic framework in an automated way. The second is that multiple performance metrics may be required to guide this automatic design process. In the first part of this work, we extend our previously proposed algorithmic framework, increasing the number of MOEAs, underlying evolutionary algorithms, and search paradigms that it comprises. These components can be combined following a general MOEA template, and an automatic configuration method is used to instantiate high-performing MOEA designs that optimize a given performance metric and present state-of-the-art performance. In the second part, we propose a multiobjective formulation for the automatic MOEA design, which proves critical for the context of many-objective optimization due to the disagreement of established performance metrics. Our proposed formulation leads to an automatically designed MOEA that presents state-of-the-art performance according to a set of metrics, rather than a single one.

Từ khóa


Tài liệu tham khảo

10.1109/CEC.2002.1007033

10.1109/CEC.2001.934295

10.1287/mnsc.25.1.73

10.1162/EVCO_a_00009

10.1016/j.ejor.2006.08.008

10.1109/TEVC.2015.2474158

10.1162/evco_a_00217

10.1007/978-3-642-00483-4

Biscani F., 2010, 4th International Conference on Astrodynamics Tools and Techniques

10.1007/3-540-36970-8_35

10.1007/978-3-319-50349-3_3

10.1023/B:HEUR.0000026900.92269.ec

10.1016/j.sorms.2015.08.001

Coello Coello C. A., 2007, Evolutionary algorithms for solving multi-objective problems

10.1007/BF01197559

Deb K., 2001, Multi-objective optimization using evolutionary algorithms

10.1504/IJAISC.2014.059280

10.1109/TEVC.2013.2281535

10.1109/4235.996017

10.1007/1-84628-137-7_6

10.1145/1570256.1570301

10.1145/2001576.2001847

10.1109/MHS.1995.494215

10.1007/BF02578918

10.1007/s10732-014-9275-9

10.1007/978-3-540-31880-4_2

Fonseca C. M., 1993, ICGA, 416

Goldberg D. E., 1989, Genetic algorithms in search, optimization and machine learning

10.1162/EVCO_a_00075

Hoos H. H., 2012, Autonomous search, 37

10.1109/TEVC.2005.861417

10.1109/CEC.2003.1299427

10.1162/evco.2007.15.1.1

Igel C., 2008, Journal of Machine Learning Research, 9:993

10.1109/TCYB.2014.2307319

KhudaBukhsh A. R., 2009, Proceedings of the Twenty-First International Joint Conference on Artificial Intelligence, 517

10.1162/106365600568167

10.1109/CEC.2005.1554717

10.1145/2792984

López-Ibáñez M., 2016, Operations Research Perspectives, 3:43

10.1007/978-3-642-19893-9_4

10.1109/TEVC.2011.2182651

10.1109/CEC.2002.1004404

10.1007/978-3-540-68830-3_7

10.1201/9781420010749.ch29

Price K., 2005, Differential evolution: A practical approach to global optimization

10.1007/978-3-540-31880-4_36

10.1007/0-387-28356-0_17

10.1109/TEVC.2010.2064321

10.1145/2001576.2001667

Tanabe R., 2017, IEEE Access, 5:19597

10.1007/978-3-540-70928-2_22

Voß T., 2010, Proceedings of the Genetic and Evolutionary Computation Conference (GECCO), 487, 10.1145/1830483.1830573

10.1109/TEVC.2007.892759

10.1109/CEC.2009.4982949

10.1007/978-3-540-30217-9_84

Zitzler E., 2002, Evolutionary Methods for Design, Optimisation and Control, 95

10.1109/TEVC.2009.2016569

10.1109/TEVC.2003.810758

Thông tin
Thông tin xuất bản

Evolutionary Computation

Tập 28 Số 2

195-226

Thông tin tác giả