Abstract
In this article, a competitive functional link artificial neural network (C-FLANN) is proposed for function approximation and classification problems. In contrast to the traditional functional link artificial neural networks (FLANNs), the novel structure is a universal approximator and can be used for various applications. C-FLANN is a single-layered feed-forward neural network that enjoys from the concepts of expanded inputs, information capacity units (ICUs) and a winner-take-all competition among the ICUs. These features increase the information capacity of the model without adding the hidden neurons. In the experimental studies, the proposed method is tested on function approximation problems as well as classification applications. Various comparisons with related algorithms such as improved swarm optimization-based FLANN, random vector FLANN and a multilayer perceptron indicate the superiority of the approach in terms of higher accuracy.
Similar content being viewed by others
Explore related subjects
Discover the latest articles, news and stories from top researchers in related subjects.References
Amin MF, Savitha R, Amin MI, Murase K (2012) Orthogonal least squares based complex-valued functional link network. Neural Netw 32:257–266
Anish CM, Majhi B (2016) Hybrid nonlinear adaptive scheme for stock market prediction using feedback FLANN and factor analysis. J Korean Stat Soc 45(1):64–76
Arakawa M, Nakayama H, Yun YB, Ishikawa H (2000) Optimum design using radial basis function networks by adaptive range genetic algorithms (determination of radius in radial basis function networks). In: Industrial electronics society, 2000. IECON 2000. 26th annual conference of the IEEE (vol 2). IEEE, pp 1219–1224
Behera SK, Das DP, Subudhi B (2014) Functional link artificial neural network applied to active noise control of a mixture of tonal and chaotic noise. Appl Soft Comput 23:51–60
Benala TR, Chinnababu K, Mall R, Dehuri S (2013) A particle swarm optimized functional link artificial neural networks (PSOFLANN) in software cost estimation. In: Proceedings of the international conference on frontiers of intelligent computing: theory and applications (FICTA) advances in intelligent systems and computing 199: 59–66
Benala TR, Dehuri S, Mall R, Dehuri S, Prasanthi VL (2012) Software effort prediction using fuzzy clustering and functional link artificial neural networks. Lect Notes Comput Sci SEMCCO 7677:124–132
Broomhead DS, Lowe D (1998) Multivariable functional interpolation and adaptive networks. Complex Syst 2:321–355
Brouwer RK (2005) Automatic training of a min-max neural network for function approximation by using a second feed forward network. Soft Comput 9(5):393–397
Carini A, Sicuranza GL (2012) A new class of FLANN filters with application to nonlinear active noise control. In: Signal processing conference (EUSIPCO), 2012 proceedings of the 20th European. IEEE, pp 1950–1954
Chakravarty S, Dash PK (2012) A PSO based integrated functional link net and interval type-2 fuzzy logic system for predicting stock market indices. Appl Soft Comput 12(2):931–941
Chandra B, Gupta M (2013) A novel approach for distance-based semi-supervised clustering using functional link neural network. Soft Comput 17(3):369–379
Comminiello D, Scarpiniti M, Azpicueta-Ruiz LA, Arenas-García J, Uncini A (2014) Nonlinear acoustic echo cancellation based on sparse functional link representations. IEEE/ACM Trans Audio Speech Lang Process 22(7):1172–1183
Comminiello D, Scarpiniti M, Scardapane S, Parisi R, Uncini A (2015) Improving nonlinear modeling capabilities of functional link adaptive filters. Neural Netw 69:51–59
Costarelli D, Spigler R (2015) Approximation by series of sigmoidal functions with applications to neural networks. Annali di Matematica Pura ed Applicata (1923-) 194(1):289–306
Cui M, Liu H, Li Z, Tang Y, Guan X (2014) Identification of Hammerstein model using functional link artificial neural network. Neurocomputing 142:419–428
Cybenko G (1989) Approximation by superpositions of a sigmoidal function. Math Control Signals Syst 2(4):303–314
Dai W, Liu Q, Chai T (2015) Particle size estimate of grinding processes using random vector functional link networks with improved robustness. Neurocomputing 169:361–372
Das DP, Panda G (2004) Active mitigation of nonlinear noise processes using a novel filtered-s LMS algorithm. IEEE Trans Speech Audio Process 12(3):313–322
Das SK, Suman S (2015) Prediction of lateral load capacity of pile in clay using multivariate adaptive regression spline and functional network. Arab J Sci Eng 40(6):1565–1578
Dash SK, Bisoi R, Dash PK (2016) A hybrid functional link dynamic neural network and evolutionary unscented Kalman filter for short-term electricity price forecasting. Neural Comput Appl 27(7):2123–2140
Dehuri S, Roy R, Cho SB, Ghosh A (2012) An improved swarm optimized functional link artificial neural network (ISO-FLANN) for classification. J Syst Softw 85(6):1333–1345
Dehuri S, Cho SB (2010) A comprehensive survey on functional link neural networks and an adaptive PSO-BP learning for CFLNN. Neural Comput Appl 19(2):187–205
Dehuri S, Cho SB (2010) A hybrid genetic based functional link artificial neural network with a statistical comparison of classifiers over multiple datasets. Neural Comput Appl 19(2):317–328
Dehuri S, Mishra BB, Cho SB (2008) Genetic feature selection for optimal functional link artificial neural network in classification. In: International conference on intelligent data engineering and automated learning. Springer, Berlin, pp 156–163
Ebrahimzadeh A, Ghazalian R (2010) Modulation classification using genetic algorithm and radial basis neural network based on the HOS. In: 2010 6th international conference on digital content, multimedia technology and its applications (IDC). IEEE, pp 375–378
Fard SP, Zainuddin Z (2015) Almost everywhere approximation capabilities of double Mellin approximate identity neural networks. Soft Comput 1–9. doi:10.1007/s00500-015-1753-y
Fu X, Wang L (20020 A GA-based RBF classifier with class-dependent features. In: Proceedings of the 2002 congress on evolutionary computation, 2002. CEC’02, vol 2. IEEE, pp. 1890–1894
Gaurav K, Mishra SK (2015) Nonlinear system identification using clonal particle swarm optimization-based functional link artificial neural network. In: Computational vision and robotics. Springer India, pp 89–96
Ghazali R, Bakar ZA, Hassim YMM, Herawan T, Wahid N (2014) Functional link neural network with modified cuckoo search training algorithm for physical time series forecasting. In: International conference on intelligent computing. Springer International Publishing, pp 285–291
Goyal V, Deolia VK, Sharma TN (2015) Robust sliding mode control for nonlinear discrete-time delayed systems based on neural network. Intell Control Autom 6(01):75
Hassim YMM, Ghazali R (2016) Improving functional link neural network learning scheme for mammographic classification. In: Advances in neural networks. Springer International Publishing, pp 213–221
Hassoun MH (1995) Fundamentals of artificial neural networks. MIT Press, Cambridge
Hernández-Aguirre A, Koutsougeras C, Buckles B (2002) Sample complexity for function learning tasks through linear neural networks. Int J Artif Intell Tools 11(04):499–511
Hornik K (1991) Approximation capabilities of multilayer feedforward networks. Neural Netw 4(2):251–257
Hsu CF (2013) Adaptive functional-link-based neural fuzzy controller design for a DC gear motor driver. Neural Comput Appl 23(1):303–313
Jia Y, Meng K, Xu Z (2015) Nk induced cascading contingency screening. IEEE Trans Power Syst 30(5):2824–2825
Kaushik A, Soni AK, Soni R (2016) An improved functional link artificial neural networks with intuitionistic fuzzy clustering for software cost estimation. Int J Syst Assur Eng Manag 7(Suppl 1):50. doi:10.1007/s13198-014-0298-2
Kecman V (2001) Learning and soft computing: support vector machines, neural networks, and fuzzy logic models. MIT Press, Cambridge
Kosko B (1991) Stochastic competitive learning. IEEE Trans Neural Netw 2(5):522–529
Li M, Liu J, Jiang Y, Feng W (2012) Complex-Chebyshev functional link neural network behavioral model for broadband wireless power amplifiers. IEEE Trans Microw Theory Tech 60(6):1979–1989
Lin FJ, Teng LT, Lin JW, Chen SY (2009) Recurrent functional-link-based fuzzy-neural-network-controlled induction-generator system using improved particle swarm optimization. IEEE Trans Ind Electron 56(5):1557–1577
Lowe D (1989). Adaptive radial basis function nonlinearities, and the problem of generalisation. In: First IEE international conference on artificial neural networks, 1989, (Conf Publ No 313). IET, pp 171–175
Mall S, Chakraverty S (2016a) Hermite functional link neural network for solving the Van der Pol–duffing oscillator equation. Neural Comput 28(8):1574–1598. doi:10.1162/NECO_a_00858
Mall S, Chakraverty S (2016b) Application of legendre neural network for solving ordinary differential equations. Appl Soft Comput 43:347–356
Martínez-Villena JM, Rosado-Muñoz A, Soria-Olivas E (2014) Hardware implementation methods in random vector functional-link networks. Appl Intell 41(1):184–195
Naik B, Nayak J, Behera HS, Abraham A (2016) A self adaptive harmony search based functional link higher order ANN for non-linear data classification. Neurocomputing 179:69–87
Nayak SK, Nayak SC, Behera HS (2016) Evolving low complex higher order neural network based classifiers for medical data classification. In: Computational intelligence in data mining, vol 2. Springer India, pp 415–425
Pao Yoh-Han (1989) Adaptive pattern recognition and neural networks. Addison-Wesley Longman Publishing Co., Inc., Boston
Pao YH, Park GH, Sobajic DJ (1994) Learning and generalization characteristics of the random vector functional-link net. Neurocomputing 6(2):163–180
Parhi P, Mishra D, Mishra S, Shaw K (2012) A novel PSO-FLANN framework of feature selection and classification for microarray data. Proc Eng 38:1644–1649
Patra A, Das S, Mishra SN, Senapati MR (2015) An adaptive local linear optimized radial basis functional neural network model for financial time series prediction. Neural Comput Appl 1–10. doi:10.1007/s00521-015-2039-0
Porwal A, Carranza EJM, Hale M (2003) Artificial neural networks for mineral-potential mapping: a case study from Aravalli Province, Western India. Nat Resour Res 12(3):155–171
Scardapane S, Comminiello D, Scarpiniti M, Uncini A (2016) A semi-supervised random vector functional-link network based on the transductive framework. Inf Sci 364:156–166
Sicuranza GL, Carini A (2012) On the BIBO stability condition of adaptive recursive FLANN filters with application to nonlinear active noise control. IEEE Trans Audio Speech Lang Process 20(1):234–245
Wang L (1997) On competitive learning. IEEE Trans Neural Netw 8(5):1214–1217
Wang L, Fu X (2006) Data mining with computational intelligence. Springer Science & Business Media, Berlin
Weng WD, Yang CS, Lin RC (2007) A channel equalizer using reduced decision feedback Chebyshev functional link artificial neural networks. Inf Sci 177(13):2642–2654
Yang S, Ting TO, Man KL, Guan SU (2013) Investigation of neural networks for function approximation. Proc Comput Sci 17:586–594
Zhang Z, Zheng N, Wang T (2001) Fuzzy generalization of the counter-propagation neural network: a family of soft competitive basis function neural networks. Soft Comput 5(6):440–450
Zhang W, Liu G, Dai H (2008) Simulation of food intake dynamics of holometabolous insect using functional link artificial neural network. Stoch Environ Res Risk Assess 22(1):123–133
Zhao H, Zhang J (2010) Pipelined Chebyshev functional link artificial recurrent neural network for nonlinear adaptive filter. IEEE Trans Syst Man Cybern Part B (Cybernetics) 40(1):162–172
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflict of interest
The authors declare that they have no conflict of interest.
Additional information
Communicated by V. Loia.
Appendix 1
Appendix 1
1.1 Universal Approximation Theorem (Cybenko 1989; Hornik 1991; Hassoun 1995)
Let \(f(\cdot )\) be a nonconstant, bounded and monotonically increasing continuous function. Let denote the n-dimensional unit hypercube \([0,1]^{m}\). The space of continuous functions on \(I^{m}\) is denoted by \(C(I^{m})\). Then, given any function \(g\in C(I^{m})\) and \(\epsilon > 0\), there exist an integer k and real constants \(\rho _i\), \(\theta _i \in \mathfrak {R}\), \(\zeta _i \in \mathfrak {R}^{m}\), where \(i=1\ldots m\) such that we may define:
as an approximate realization of the function g, where G is independent of f, that is,
for all \(\vec {X}\in I^{m}\). In other words, functions of the form \(G(\vec {X})\) are dense in \(C(I^{m})\).
Rights and permissions
About this article
Cite this article
Lotfi, E., Rezaee, A.A. A competitive functional link artificial neural network as a universal approximator. Soft Comput 22, 4613–4625 (2018). https://doi.org/10.1007/s00500-017-2644-1
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00500-017-2644-1