Abstract
Determining appropriate weights and biases for feed-forward neural networks is a critical task. Despite the prevalence of gradient-based methods for training, these approaches suffer from sensitivity to initial values and susceptibility to local optima. To address these challenges, we introduce a novel two-level clustering-based differential evolution approach, C2L-DE, to identify the initial seed for a gradient-based algorithm. In the initial phase, clustering is employed to detect some regions in the search space. Population updates are then executed based on the information available within each region. A new central point is proposed in the subsequent phase, leveraging cluster centres for incorporation into the population. Our C2L-DE algorithm is compared against several recent DE-based neural network training algorithms, and is shown to yield favourable performance.
References
Abrishami, N., Sepaskhah, A.R., Shahrokhnia, M.H.: Estimating wheat and maize daily evapotranspiration using artificial neural network. Theor. Appl. Climatol. 135(3), 945–958 (2018). https://doi.org/10.1007/s00704-018-2418-4
Amirsadri, S., Mousavirad, S.J., Ebrahimpour-Komleh, H.: A levy flight-based grey wolf optimizer combined with back-propagation algorithm for neural network training. Neural Comput. Appl. 30(12), 3707–3720 (2017). https://doi.org/10.1007/s00521-017-2952-5
Bojnordi, E., Mousavirad, S.J., Pedram, M., et al.: Improving the generalisation ability of neural networks using a Lévy flight distribution algorithm for classification problems. New Gener. Comput. 41(2), 225–242 (2023). https://doi.org/10.1007/s00354-023-00214-5
Bojnordi,E., Mousavirad,S.J., Schaefer, G., Korovin, I.: MCS-HMS: a multi-cluster selection strategy for the human mental search algorithm. In: IEEE Symposium Series on Computational Intelligence, pp. 1–6, 2021
Cai, Z., Gong, W., Ling, C.X., Zhang, H.: A clustering-based differential evolution for global optimization. Appl. Soft Comput. 11(1), 1363–1379 (2011)
Deb, K.: A population-based algorithm-generator for real-parameter optimization. Soft. Comput. 9(4), 236–253 (2005)
Ding, S., Chunyang, S., Junzhao, Yu.: An optimizing BP neural network algorithm based on genetic algorithm. Artif. Intell. Rev. 36(2), 153–162 (2011)
Duan, H., Huang, L.: Imperialist competitive algorithm optimized artificial neural networks for UCAV global path planning. Neurocomputing 125, 166–171 (2014)
El-Bakry, M.Y., El-Dahshan, E.-S.A., El-Hamied, E.F.A.: Charged particle pseudorapidity distributions for Pb-Pb and Au-Au collisions using neural network model. Ukrainian J. Phys. 58(8), 709–709 (2013)
Fister, I., Fister, D., Deb, S., Mlakar, U., Brest, J.: Post hoc analysis of sport performance with differential evolution. Neural Comput. Appl. 32, 1–10 (2018)
Hosaka, T.: Bankruptcy prediction using imaged financial ratios and convolutional neural networks. Expert Syst. Appl. 117, 287–299 (2019)
Ilonen, J., Kamarainen, J.-K., Lampinen, J.: Differential evolution training algorithm for feed-forward neural networks. Neural Process. Lett. 17(1), 93–105 (2003)
Kim, D., Kim, H., Chung, D.: A modified genetic algorithm for fast training neural networks. In: International Symposium on Neural Networks, pp. 660–665 (2005)
Lera, G., Pinzolas, M.: Neighborhood based Levenberg-Marquardt algorithm for neural network training. IEEE Trans. Neural Networks 13(5), 1200–1203 (2002)
Levenberg, K.: A method for the solution of certain non-linear problems in least squares. Q. Appl. Math. 2(2), 164–168 (1944)
MacQueen, J.: Some methods for classification and analysis of multivariate observations. In: 5th Berkeley Symposium on Mathematical Statistics and Probability, pp. 281–297 (1967)
Mandal, S., Saha, G., Pal, R.K.: Neural network training using firefly algorithm. Glob. J. Adv. Eng. Sci. 1(1), 7–11 (2015)
Marquardt, D.W.: An algorithm for least-squares estimation of nonlinear parameters. J. Soc. Ind. Appl. Math. 11(2), 431–441 (1963)
Mirjalili, S.: How effective is the Grey Wolf optimizer in training multi-layer perceptrons. Appl. Intell. 43(1), 150–161 (2015). https://doi.org/10.1007/s10489-014-0645-7
Moravvej, S.V., Mousavirad, S.J., Oliva, D., Schaefer, G., Sobhaninia, Z.: An improved DE algorithm to optimise the learning process of a BERT-based plagiarism detection model. In: IEEE Congress on Evolutionary Computation, pp. 1–7 (2022)
Mousavirad, S.J., Bidgoli, A.A., Rahnamayan, S.: Tackling deceptive optimization problems using opposition-based DE with center-based Latin hypercube initialization. In: 14th International Conference on Computer Science and Education (2019)
Mousavirad, S.J., Bidgoli, A.A., Ebrahimpour-Komleh, H., G.S.: A memetic imperialist competitive algorithm with chaotic maps for multi-layer neural network training. Int. J. Bio-Inspired Comput. 14(4), 227–236 (2019)
Mousavirad, S.J., Bidgoli, A.A., Ebrahimpour-Komleh, H., Schaefer, G., Korovin, I.: An effective hybrid approach for optimising the learning process of multi-layer neural networks. In: International Symposium on Neural Networks, pp. 309–317 (2019)
Mousavirad, S.J., Gandomi, A.H., Homayoun, H.: A clustering-based differential evolution boosted by a regularisation-based objective function and a local refinement for neural network training. In: IEEE Congress on Evolutionary Computation, pp. 1–8 (2022)
Mousavirad, S.J., Jalali, S.M.J., Sajad, A., Abbas, K., Schaefer, G., Nahavandi, S.: Neural network training using a biogeography-based learning strategy. In: International Conference on Neural Information Processing (2020)
Mousavirad, S.J., Oliva, D., Hinojosa, S., Schaefer, G.: Differential evolution-based neural network training incorporating a centroid-based strategy and dynamic opposition-based learning. In: IEEE Congress on Evolutionary Computation, pp. 1233–1240 (2021)
Mousavirad, S.J., Rahmani, R., Dolatabadi, N.: A transfer learning based artificial neural network in geometrical design of textured surfaces for tribological applications. Surf. Topogr. Metrol. Prop. 11(2), 025001 (2023)
Mousavirad, S.J., Rahnamayan, S.: Evolving feedforward neural networks using a quasi-opposition-based differential evolution for data classification. In: IEEE Symposium Series on Computational Intelligence (2020)
Mousavirad, S.J., Rahnamayan, S.: A novel center-based differential evolution algorithm. In: Congress on Evolutionary Computation (2020)
Mousavirad, S.J., Schaefer, G., Jalali, S.M.J., Korovin, I.: A benchmark of recent population-based metaheuristic algorithms for multi-layer neural network training. In: Genetic and Evolutionary Computation Conference Companion, pp. 1402–1408 (2020)
Mousavirad, S.J., Schaefer, G., Korovin, I.: An effective approach for neural network training based on comprehensive learning. In: International Conference on Pattern Recognition (2020)
Mousavirad, S.J., Schaefer, G., Korovin, I., Oliva, D.: RDE-OP: a region-based differential evolution algorithm incorporation opposition-based learning for optimising the learning process of multi-layer neural networks. In: Castillo, P.A., Jiménez Laredo, J.L. (eds.) EvoApplications 2021. LNCS, vol. 12694, pp. 407–420. Springer, Cham (2021). https://doi.org/10.1007/978-3-030-72699-7_26
Munkhdalai, L., Lee, J.Y., Ryu, K.H.: A hybrid credit scoring model using neural networks and logistic regression. In: Pan, J.-S., Li, J., Tsai, P.-W., Jain, L.C. (eds.) Advances in Intelligent Information Hiding and Multimedia Signal Processing. SIST, vol. 156, pp. 251–258. Springer, Singapore (2020). https://doi.org/10.1007/978-981-13-9714-1_27
Nawi, N.M., khan, A., Rehman, M.Z., Aziz, M.A., Herawan, T., Abawajy, J.H.: An accelerated particle swarm optimization based Levenberg Marquardt back propagation algorithm. In: Loo, C.K., Yap, K.S., Wong, K.W., Teoh, A., Huang, K. (eds.) ICONIP 2014. LNCS, vol. 8835, pp. 245–253. Springer, Cham (2014). https://doi.org/10.1007/978-3-319-12640-1_30
Pedram, M., Mousavirad, S.J., Schaefer, G.: Training neural networks with Lévy flight distribution algorithm. In: 7th International Conference on Harmony Search, Soft Computing and Applications, pp. 93–103 (2022)
Rahmani, S., Mousavirad, S.J., El-Abd, M., Schaefer, G., Oliva, D.: Centroid-based differential evolution with composite trial vector generation strategies for neural network training. In: Correia, J., Smith, S., Qaddoura, R. (eds.) International Conference on the Applications of Evolutionary Computation, vol. 13989, pp. 608–622. Springer, Cham (2023). https://doi.org/10.1007/978-3-031-30229-9_39
Sexton, R.S., Gupta, J.N.D.: Comparative evaluation of genetic algorithm and backpropagation for training neural networks. Inform. Sci. 129(1–4), 45–59 (2000)
Shi, Y., Eberhart, R.: A modified particle swarm optimizer. In: IEEE International Conference on Evolutionary Computation, pp. 69–73 (1998)
Si, T., Dutta, R.: Partial opposition-based particle swarm optimizer in artificial neural network training for medical data classification. Int. J. Inform. Technol. Decis. Making 18(5), 1717–1750 (2019)
Storn, R., Price, K.: Differential evolution-a simple and efficient heuristic for global optimization over continuous spaces. J. Global Optim. 11(4), 341–359 (1997)
Wang, X., et al.: Massive expansion and differential evolution of small heat shock proteins with wheat (triticum aestivum l.) polyploidization. Sci. Rep. 7(1), 1–12 (2017)
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
Mousavirad, S.J., Oliva, D., Schaefer, G., Moghadam, M.H., El-Abd, M. (2024). A Novel Two-Level Clustering-Based Differential Evolution Algorithm for Training Neural Networks. In: Smith, S., Correia, J., Cintrano, C. (eds) Applications of Evolutionary Computation. EvoApplications 2024. Lecture Notes in Computer Science, vol 14634. Springer, Cham. https://doi.org/10.1007/978-3-031-56852-7_17
Download citation
DOI: https://doi.org/10.1007/978-3-031-56852-7_17
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-56851-0
Online ISBN: 978-3-031-56852-7
eBook Packages: Computer ScienceComputer Science (R0)