Abstract
Inspired by biological evolution’s ability to produce complex and intelligent beings, neuroevolution utilizes evolutionary algorithms for optimizing the connection weights and structure of artificial neural networks. With evolutionary algorithms often failing to produce the same level of diversity as biological evolution, explicitly encouraging diversity with additional optimization objectives has emerged as a successful approach. However, there is a lack of knowledge regarding the performance of different types of diversity objectives on problems with different characteristics. In this paper, we perform a systematic comparison between objectives related to structural diversity, behavioral diversity, and our newly proposed representational diversity. We explore these objectives’ effects on problems with different levels of modularity, regularity, deceptiveness and discreteness and find clear relationships between problem characteristics and the effect of different diversity objectives – suggesting that there is much to be gained from adapting diversity objectives to the specific problem being solved.
Source code: https://github.com/dreilstad/Neuroevolution.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Clune, J., Mouret, J.B., Lipson, H.: The evolutionary origins of modularity. Proceedings. Biological sciences / The Royal Society 280, 20122863 (07 2013). https://doi.org/10.1098/rspb.2012.2863
Cuccu, G., Gomez, F.: When novelty is not enough. In: Proceedings of the 2011 International Conference on Applications of Evolutionary Computation - Volume Part I. p. 234–243. EvoApplications’11, Springer-Verlag, Berlin, Heidelberg (2011)
Deb, K., Pratap, A., Agarwal, S., Meyarivan, T.: A fast and elitist multiobjective genetic algorithm: Nsga-ii. IEEE Trans. Evol. Comput. 6(2), 182–197 (2002). https://doi.org/10.1109/4235.996017
Deb, K.: Multi-objective optimization using evolutionary algorithms (2008)
Doncieux, S., Mouret, J.B.: Behavioral diversity measures for evolutionary robotics. In: IEEE congress on evolutionary computation. pp. 1–8. IEEE (2010)
Ellefsen, K.O., Huizinga, J., Torresen, J.: Guiding neuroevolution with structural objectives. Evol. Comput. 28(1), 115–140 (2020). https://doi.org/10.1162/evco_a_00250
Gomez, F.: Sustaining diversity using behavioral information distance. pp. 113–120 (01 2009). https://doi.org/10.1145/1569901.1569918
Griffiths, T.D., Ekárt, A.: Improving the tartarus problem as a benchmark in genetic programming. In: McDermott, J., Castelli, M., Sekanina, L., et al. (eds.) Genetic programming, pp. 278–293. Lecture Notes in Computer Science, Springer, NLD (March (2017)
Huizinga, J., Mouret, J.B., Clune, J.: Does Aligning Phenotypic and Genotypic Modularity Improve the Evolution of Neural Networks? In: Proceedings of the 25th Genetic and Evolutionary Computation Conference (GECCO). pp. 125–132. ACM, Denver, France (2016). https://doi.org/10.1145/2908812.2908836
Kashtan, N., Alon, U.: Spontaneous evolution of modularity and network motifs. Proc. Natl. Acad. Sci. 102(39), 13773–13778 (2005). https://doi.org/10.1073/pnas.0503610102
Kornblith, S., Norouzi, M., Lee, H., Hinton, G.: Similarity of neural network representations revisited. In: International Conference on Machine Learning. pp. 3519–3529. PMLR (2019)
Krčah, P.: Solving deceptive tasks in robot body-brain co-evolution by searching for behavioral novelty. In: 2010 10th International Conference on Intelligent Systems Design and Applications. pp. 284–289 (2010). https://doi.org/10.1109/ISDA.2010.5687250
LeCun, Y., Bengio, Y., Hinton, G.: Deep learning. Nature 521(7553), 436–444 (2015). https://doi.org/10.1038/nature14539
Lehman, J., Chen, J., Clune, J., Stanley, K.O.: Safe mutations for deep and recurrent neural networks through output gradients. CoRR abs/1712.06563 (2017)
Lehman, J., Stanley, K.: Abandoning objectives: Evolution through the search for novelty alone. Evolutionary computation 19, 189–223 (06 2011). https://doi.org/10.1162/EVCO_a_00025
Lehman, J., Stanley, K.O.: Evolving a diversity of virtual creatures through novelty search and local competition. In: Proceedings of the 13th Annual Conference on Genetic and Evolutionary Computation. p. 211–218. GECCO ’11, Association for Computing Machinery, New York, NY, USA (2011). https://doi.org/10.1145/2001576.2001606
Lehman, J., Stanley, K.O., Miikkulainen, R.: Effective diversity maintenance in deceptive domains. In: Proceedings of the 15th Annual Conference on Genetic and Evolutionary Computation. p. 215–222. GECCO ’13, Association for Computing Machinery, New York, NY, USA (2013). https://doi.org/10.1145/2463372.2463393
Li, J., Storie, J., Clune, J.: Encouraging creative thinking in robots improves their ability to solve challenging problems. In: Proceedings of the 2014 Annual Conference on Genetic and Evolutionary Computation. p. 193–200. GECCO ’14, Association for Computing Machinery, New York, NY, USA (2014). https://doi.org/10.1145/2576768.2598222
Mountcastle, V.B.: The columnar organization of the neocortex. Brain: a journal of neurology 120(4), 701–722 (1997)
Mouret, J.B.: Novelty-Based Multiobjectivization, vol. 341, pp. 139–154 (02 2011). https://doi.org/10.1007/978-3-642-18272-3_10
Mouret, J.B., Clune, J.: Illuminating search spaces by mapping elites. ArXiv abs/1504.04909 (2015)
Mouret, J.B., Doncieux, S.: Using Behavioral Exploration Objectives to Solve Deceptive Problems in Neuro-evolution. In: The 11th Annual conference on Genetic and evolutionary computation (GECCO’09). pp. 627–634. ACM, Montréal, Canada (2009). https://doi.org/10.1145/1569901.1569988
Mouret, J.B., Doncieux, S.: Encouraging Behavioral Diversity in Evolutionary Robotics: an Empirical Study. Evol. Comput. 20(1), 91–133 (2012). https://doi.org/10.1162/EVCO_a_00048
Reilstad, D.S.: Cultivating Diversity: a Comparison of Diversity Objectives in Neuroevolution. Master’s thesis, University of Oslo (2023), https://www.duo.uio.no/handle/10852/103916
Rothlauf, F., Rothlauf, F.: Representations for genetic and evolutionary algorithms. Springer (2006)
Stanley, K., Clune, J., Lehman, J., Miikkulainen, R.: Designing neural networks through neuroevolution. Nature Machine Intelligence 1 (01 2019). https://doi.org/10.1038/s42256-018-0006-z
Striedter, G.F.: Principles of brain evolution. Sinauer associates (2005)
Toffolo, A., Benini, E.: Genetic diversity as an objective in multi-objective evolutionary algorithms. Evol. Comput. 11(2), 151–167 (may 2003). https://doi.org/10.1162/106365603766646816
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2024 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Reilstad, D.S., Ellefsen, K.O. (2024). Cultivating Diversity: A Comparison of Diversity Objectives in Neuroevolution. In: Smith, S., Correia, J., Cintrano, C. (eds) Applications of Evolutionary Computation. EvoApplications 2024. Lecture Notes in Computer Science, vol 14635. Springer, Cham. https://doi.org/10.1007/978-3-031-56855-8_2
Download citation
DOI: https://doi.org/10.1007/978-3-031-56855-8_2
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-56854-1
Online ISBN: 978-3-031-56855-8
eBook Packages: Computer ScienceComputer Science (R0)