Abstract
In this paper, convergence results in a multivariate setting have been proved for a family of neural network operators of the max-product type. In particular, the coefficients expressed by Kantorovich type means allow to treat the theory in the general frame of the Orlicz spaces, which includes as particular case the \(L^p\)-spaces. Examples of sigmoidal activation functions are discussed, for the above operators in different cases of Orlicz spaces. Finally, concrete applications to real-world cases have been presented in both univariate and multivariate settings. In particular, the case of reconstruction and enhancement of biomedical (vascular) image has been discussed in detail.
Similar content being viewed by others
References
Adler A, Guardo R (1994) A neural network image reconstruction technique for electrical impedance tomography. IEEE Trans Med Imaging 13(4):594–600
Angeloni L, Costarelli D, Vinti G (2018) A characterization of the convergence in variation for the generalized sampling series. Annales Academiae Scientiarum Fennicae Mathematica 43:755–767
Angeloni L, Vinti G (2009) Convergence and rate of approximation for linear integral operators in \(BV^\varphi \)-spaces in multidimensional setting. J Math Anal Appl 349:317–334
Angeloni L, Vinti G (2010) Approximation with respect to Goffman–Serrin variation by means of non-convolution type integral operators. Numer Funct Anal Optim 31:519–548
Angeloni L, Vinti G (2014) Convergence and rate of approximation in BV\(\varphi ({\mathbb{R}}^N_+)\) for a class of Mellin integral operators. Atti Accad Naz Lincei Rend Lincei Mat Appl 25:217–232
Asdrubali F, Baldinelli G, Bianchi F, Costarelli D, Rotili A, Seracini M, Vinti G (2018) Detection of thermal bridges from thermographic images by means of image processing approximation algorithms. Appl Math Comput 317:160–171
Asdrubali F, Baldinelli G, Bianchi F, Costarelli D, Evangelisti L, Rotili A, Seracini M, Vinti G (2018) A model for the improvement of thermal bridges quantitative assessment by infrared thermography. Appl Energy 211:854–864
Ball KR, Grant C, Mundy WR, Shafera TJ (2017) A multivariate extension of mutual information for growing neural networks. Neural Netw 95:29–43
Bardaro C, Karsli H, Vinti G (2011) Nonlinear integral operators with homogeneous kernels: pointwise approximation theorems. Appl Anal 90(3–4):463–474
Bardaro C, Musielak J, Vinti G (2003) Nonlinear integral operators and applications. Series in nonlinear analysis and applications 9. W. De Gruyter & Co., Berlin
Bartoccini B, Costarelli D, Vinti G (2018) Extension of saturation theorems for the sampling Kantorovich operators. Complex Anal Oper Theory. https://doi.org/10.1007/s11785-018-0852-z
Bede B, Coroianu L, Gal SG (2016) Approximation by max-product type operators. Springer International Publishing, Berlin. https://doi.org/10.1007/978-3-319-34189-7
Boccuto A, Candeloro D, Sambucini AR (2017) \(L^p\) spaces in vector lattices and applications. Math Slov 67(6):1409–1426. https://doi.org/10.1515/ms-2017-0060
Bono-Nuez A, Bernal-Ruíz C, Martín-del-Brío B, Pérez-Cebolla FJ, Martínez-Iturbe A (2017) Recipient size estimation for induction heating home appliances based on artificial neural networks. Neural Comput Appl 28(11):3197–3207
Candeloro D, Sambucini AR (2015) Filter convergence and decompositions for vector lattice-valued measures. Mediterr J Math 12(3):621–637. https://doi.org/10.1007/s00009-014-0431-0
Cao F, Chen Z (2009) The approximation operators with sigmoidal functions. Comput Math Appl 58(4):758–765
Cao F, Chen Z (2012) The construction and approximation of a class of neural networks operators with ramp functions. J Comput Anal Appl 14(1):101–112
Cao F, Liu B, Park DS (2013) Image classification based on effective extreme learning machine. Neurocomputing 102:90–97
Cheang GHL (2010) Approximation with neural networks activated by ramp sigmoids. J Approx Theory 162:1450–1465
Coroianu L, Gal SG (2014) Saturation and inverse results for the Bernstein max-product operator. Period Math Hungar 69:126–133
Coroianu L, Gal SG (2017) \(L^p\)-approximation by truncated max-product sampling operators of Kantorovich-type based on Fejer kernel. J Integral Equ Appl 29(2):349–364
Costarelli D, Minotti AM, Vinti G (2017) Approximation of discontinuous signals by sampling Kantorovich series. J Math Anal Appl 450(2):1083–1103
Costarelli D, Sambucini AR (2018) Approximation results in Orlicz spaces for sequences of Kantorovich max-product neural network operators. Results Math 73(1):15. https://doi.org/10.1007/s00025-018-0799-4
Costarelli D, Spigler R (2015) How sharp is the Jensen inequality? J Inequal Appl 69:1–10
Costarelli D, Spigler R (2018) Solving numerically nonlinear systems of balance laws by multivariate sigmoidal functions approximation. Comput Appl Math 37(1):99–133
Costarelli D, Vinti G (2016) Approximation by max-product neural network operators of Kantorovich type. Results Math 69(3):505–519
Costarelli D, Vinti G (2016) Max-product neural network and quasi-interpolation operators activated by sigmoidal functions. J Approx Theory 209:1–22
Costarelli D, Vinti G (2016) Pointwise and uniform approximation by multivariate neural network operators of the max-product type. Neural Netw 81:81–90
Costarelli D, Vinti G (2017) Saturation classes for max-product neural network operators activated by sigmoidal functions. Results Math 72(3):1555–1569
Costarelli D, Vinti G (2017) Convergence for a family of neural network operators in Orlicz spaces. Mathematische Nachrichten 290(2–3):226–235
Costarelli D, Vinti G (2017) Convergence results for a family of Kantorovich max-product neural network operators in a multivariate setting. Mathematica Slovaca 67(6):1469–1480
Costarelli D, Vinti G (2018) Estimates for the neural network operators of the max-product type with continuous and p-integrable functions. Results Math 73(1):12. https://doi.org/10.1007/s00025-018-0790-0
Costarelli D, Vinti G (2018) An inverse result of approximation by sampling Kantorovich series. Proc Edinb Math Soc. https://doi.org/10.1017/S0013091518000342
Egmont-Petersena M, de Ridderb D, Handels H (2002) Image processing with neural networks—a review. Pattern Recognit 35:2279–2301
Gnecco G (2012) A comparison between fixed-basis and variable-basis schemes for function approximation and functional optimization. J Appl Math. https://doi.org/10.1155/2012/806945
Gnecco G, Sanguineti M (2011) On a variational norm tailored to variable-basis approximation schemes. IEEE Trans Inf Theory 57:549–558
Goh ATC (1995) Back-propagation neural networks for modeling complex systems. Artif Intell Eng 9:143–151
Gotleyb D, Lo Sciuto G, Napoli C, Shikler R, Tramontana E, Wozniak M (2016) Characterization and modeling of organic solar cells by using radial basis neural networks. Artif Intell Soft Comput. https://doi.org/10.1007/978-3-319-39378-0_9
Guliyev NJ, Ismailov VE (2018) On the approximation by single hidden layer feedforward neural networks with fixed weights. Neural Networks 98:296–304
Guliyev NJ, Ismailov VE (2018) Approximation capability of two hidden layer feedforward neural networks with fixed weights. Neurocomputing 316:262–269
Iliev A, Kyurkchiev N, Markov S (2015) On the approximation of the cut and step functions by logistic and Gompertz functions. Biomath. https://doi.org/10.11145/j.biomath.2015.10.101
Kainen PC, Kurkova V, Sanguineti M (2009) Complexity of Gaussian-radial-basis networks approximating smooth functions. J Complex 25(1):63–74
Lai G, Liu Z, Zhang Y, Philip Chen CL (2016) Adaptive position/attitude tracking control of aerial robot with unknown inertial matrix based on a new robust neural identifier. IEEE Trans Neural Netw Learn Syst 27(1):18–31
Liu P, Wang J, Zeng Z (2017) Multistability of delayed recurrent neural networks with Mexican hat activation functions. Neural Comput 29(2):423–457
Livingstone DJ (2008) Artificial neural networks: methods and applications (methods in molecular biology). Humana Press, New York
Maiorov V (2006) Approximation by neural networks and learning theory. J Complex 22(1):102–117
Musielak J (1983) Orlicz spaces and modular spaces. Lecture notes in mathematics, vol 1034. Springer, Berlin
Musielak J, Orlicz W (1959) On modular spaces. Studia Math 28:49–65
Olivera JJ (2017) Global exponential stability of nonautonomous neural network models with unbounded delays. Neural Netw 96:71–79
Rister B, Rubin DL (2017) Piecewise convexity of artificial neural networks. Neural Netw 94:34–45
Sahoo A, Xu H, Jagannathan S (2016) Adaptive neural network-based event-triggered control of single-input single-output nonlinear discrete-time systems. IEEE Trans Neural Netw Learn Syst 27(1):151–164
Stamov G, Stamova I (2017) Impulsive fractional-order neural networks with time-varying delays: almost periodic solutions. Neural Comput Appl 28(11):3307–3316
Acknowledgements
The authors would like to thank the referees for their useful suggestions which led us to insert the section devoted to real-world applications.
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflict of interest
The authors declare that they have no conflict of interest.
Ethical statement
Ethical approval was waived considering that the CT images analyzed were anonymized and the results did not influence any clinical judgment.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
The authors are members of the Gruppo Nazionale per l’Analisi Matematica, la Probabilità e le loro Applicazioni (GNAMPA) of the Istituto Nazionale di Alta Matematica (INdAM). Moreover, the first and the second authors of the paper have been partially supported within the 2018 GNAMPA-INdAM Project “Dinamiche non autonome, analisi reale e applicazioni,” while the second and the third authors within the project: Ricerca di Base 2017 dell’Università degli Studi di Perugia—“Metodi di teoria degli operatori e di Analisi Reale per problemi di approssimazione ed applicazioni.”
Rights and permissions
About this article
Cite this article
Costarelli, D., Sambucini, A.R. & Vinti, G. Convergence in Orlicz spaces by means of the multivariate max-product neural network operators of the Kantorovich type and applications. Neural Comput & Applic 31, 5069–5078 (2019). https://doi.org/10.1007/s00521-018-03998-6
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00521-018-03998-6
Keywords
- Sigmoidal function
- Multivariate max-product neural network operator
- Orlicz space
- Modular convergence
- Neurocomputing process
- Data modeling
- Image processing