Abstract
Estimating a form of equation that explains data is very useful to understand various physical, chemical, social, and biological phenomena. One effective approach for finding the form of an equation is to solve the symbolic regression problem using genetic programming (GP). However, this approach requires a long computation time because of the explosion of the number of combinations of candidate functions that are used as elements to construct equations. In the present paper, a novel method to effectively eliminate unnecessary functions from an initial set of functions using a deep neural network was proposed to reduce the number of computations of GP. Moreover, a method was proposed to improve the accuracy of the classification using eigenvalues when classifying whether functions are required for symbolic regression. Experiment results showed that the proposed method can successfully classify functions with over 90% of the data created in the present study.
Keywords
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsReferences
Schmidt, M.D., Lipson, H.: Distilling free-form natural laws from experimental data. Science 324(5923), 81–85 (2009)
Schmidt, M.D., Lipson, H.: Coevolution of fitness predictors. IEEE Trans. Evol. Comput. 12(6), 736–749 (2008)
Martius, G., Lampert, C.H.: Extrapolation and learning equations. CoRR, arXiv:1610.02995 (2016)
Van Hemert, E., Eggermont, J., Hemert, J.I.: Stepwise adaptation of weights for symbolic regression with genetic programming. In: Proceedings of the Twelveth Belgium/Netherlands Conference on Arti Intelligence (BNAIC 2000) (2000)
Schmidt, M.D., Lipson, H.: Co-evolution of fitness maximizers and fitness predictors. In: GECCO Late Breaking Paper (2005)
Schmidt, M.D., Lipson, H.: Co-evolving fitness predictors for accelerating and reducing evaluations. Genet. Program. Theor. Pract. IV 5, 113–130 (2006)
Uy, N.Q., Hoai, N.X., O’Neill, M., McKay, R.I., Galván-López, E.: Semantically-based crossover in genetic programming: application to real-valued symbolic regression. Genet. Program. Evolvable Mach. 12(2), 91–119 (2011)
Suzuki, I., Ikeuchi, Y., Tsumura, T., Nakashima, Y., Nakashima, H.: A speedup technique for GA with reuse. In: IPSJ Transactions on Advanced Computing Systems, vol. 46, no. SIG 16(ACS 12), pp. 129–143, December 2005
Haeri, M.A., Ebadzadeh, M.M., Folino, G.: Statistical genetic programming for symbolic regression. Appl. Soft Comput. 60, 447–469 (2017)
Maronna, R., Lovric, M.: Robust statistical methods. In: Lovric, M. (ed.) International Encyclopedia of Statistical Science, pp. 1244–1248. Springer, Heidelberg (2011). https://doi.org/10.1007/978-3-642-04898-2
Kingma, D.P., Ba, J.L.: Adam: A Method for Stochastic Optimization. CoRR, arXiv:1412.6980 (2014)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2019 Springer Nature Switzerland AG
About this paper
Cite this paper
Koga, I., Ono, K. (2019). Effective Pre-processing of Genetic Programming for Solving Symbolic Regression in Equation Extraction. In: Kotzinos, D., Laurent, D., Spyratos, N., Tanaka, Y., Taniguchi, Ri. (eds) Information Search, Integration, and Personalization. ISIP 2018. Communications in Computer and Information Science, vol 1040. Springer, Cham. https://doi.org/10.1007/978-3-030-30284-9_6
Download citation
DOI: https://doi.org/10.1007/978-3-030-30284-9_6
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-30283-2
Online ISBN: 978-3-030-30284-9
eBook Packages: Computer ScienceComputer Science (R0)