Abstract
Recently, Timotheou has formulated the learning problem of the random neural network (RNN) into a convex non-negative least-square problem that can be solved to optimality. By incorporating this work of problem formulation and the line-search technique, this paper designs a line-search aided non-negative least-square (LNNLS) learning algorithm for the RNN, which is able to find a nearly optimal solution efficiently. (The source code is available at www.yonghuayin.icoc.cc.) Numerical experiments based on datasets with different dimensions have been conducted to demonstrate the efficacy of the LNNLS learning algorithm.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Gelenbe, E.: Learning in the recurrent random neural network. Neural Comput. 5(1), 154–164 (1993)
Timotheou, S.: A novel weight initialization method for the random neural network. Neurocomputing 73(1), 160–168 (2009)
Timotheou, S.: Nonnegative least squares learning for the random neural network. In: Artificial Neural Networks-ICANN, pp. 195–204 (2005)
Gelenbe, E., Hussain, K.F.: Learning in the multiple class random neural network. IEEE Trans. Neural Netw. 13(6), 1257–1267 (2002)
Gelenbe, E., Sungur, M.: Random network learning and image compression. In: IEEE International Conference on Neural Networks, pp. 3996–3999 (1994)
Zhang, Y., Yin, Y., Guo, D., Yu, X., Xiao, L.: Cross-validation based weights and structure determination of Chebyshev-polynomial neural networks for pattern classification. Pattern Recogn. 47(10), 3414–3428 (2014)
Zhang, Y., Yang, Y., Cai, B., Guo, D.: Zhang neural network and its application to Newton iteration for matrix square root estimation. Neural Comput. Appl. 21(3), 453–460 (2012)
Diniz-Ehrhardt, M.A., Martnez, J.M., Raydn, M.: A derivative-free nonmonotone line-search technique for unconstrained optimization. J. Comput. Appl. math. 219(2), 383–397 (2008)
Lin, C.J.: Projected gradient methods for nonnegative matrix factorization. Neural comput. 19(10), 2756–2779 (2007)
Gelenbe, E.: Energy packet networks: smart electricity storage to meet surges in demand. In: 5th International ICST Conference on Simulation Tools and Techniques, pp. 1–7 (2012)
Gelenbe, E. Energy packet networks: adaptive energy management for the cloud. In: 2nd International Workshop on Cloud Computing Platforms, pp. 1 (2012)
Gelenbe, E., Labed, A.: G-networks with multiple classes of signals and positive customers. Eur. J. Oper. Res. 108(2), 293–305 (1998)
Acknowledgments
This work is funded by an Imperial College Ph.D. Scholarship.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2016 Springer International Publishing Switzerland
About this paper
Cite this paper
Yin, Y. (2016). Line-Search Aided Non-negative Least-Square Learning for Random Neural Network. In: Abdelrahman, O., Gelenbe, E., Gorbil, G., Lent, R. (eds) Information Sciences and Systems 2015. Lecture Notes in Electrical Engineering, vol 363. Springer, Cham. https://doi.org/10.1007/978-3-319-22635-4_16
Download citation
DOI: https://doi.org/10.1007/978-3-319-22635-4_16
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-22634-7
Online ISBN: 978-3-319-22635-4
eBook Packages: EngineeringEngineering (R0)