Abstract
It has been shown that, by adding a chaotic sequence to the weight update during the training of neural networks, the chaos injection-based gradient method (CIBGM) is superior to the standard backpropagation algorithm. This paper presents the theoretical convergence analysis of CIBGM for training feedforward neural networks. We consider both the case of batch learning as well as the case of online learning. Under mild conditions, we prove the weak convergence, i.e., the training error tends to a constant and the gradient of the error function tends to zero. Moreover, the strong convergence of CIBGM is also obtained with the help of an extra condition. The theoretical results are substantiated by a simulation example.
Similar content being viewed by others
References
Ahmed SU, Shahjahan M, Murase K (2011) Injecting chaos in feedforward neural networks. Neural Process Lett 34:87–100
Behera L, Kumar S, Patnaik A (2006) On adaptive learning rate that guarantees convergence in feedforward networks. IEEE Trans Neural Netw 17(5):1116–1125
Bertsekas DP, Tsitsiklis JN (2000) Gradient convergence in gradient methods with errors. SIAM J Optim 3:627–642
Charalambous C (1992) Conjugate gradient algorithm for efficient training of artificial neural networks. Inst Electr Eng Proc 139:301–310
Fan QW, Zurada JM, Wu W (2014) Convergence of online gradient method for feedforward neural networks with smoothing L1/2 regularization penalty. Neurocomputing 131:208–216
Fine TL, Mukherjee S (1999) Parameter convergence and learning curves for neural networks. Neural Comput 11:747–769
Guo DQ (2011) Inhibition of rhythmic spiking by colored noise in neural systems. Cogn Neurodyn 5(3):293–300
Hagan MT, Mehnaj MB (1994) Training feedforward networks with Marquardt algorithm. IEEE Trans Neural Netw 5(6):989–993
Haykin S (2008) Neural networks and learning machines. Prentice Hall, New Jersey
Heskes T, Wiegerinck W (1996) A theoretical comparison of batch-mode, on-line, cyclic, and almost-cyclic learning. IEEE Trans Neural Netw 7(4):919–925
Ho KI, Leung CS, Sum JP (2010) Convergence and objective functions of some fault/noise-injection-based online learning algorithms for RBF networks. IEEE Trans Neural Netw 21(6):938–947
Iiguni Y, Sakai H, Tokumaru H (1992) A real-time learning algorithm for a multilayered neural netwok based on extended Kalman filter. IEEE Trans Signal Process 40(4):959–966
Karnin ED (1990) A simple procedure for pruning back-propagation trained neural networks. IEEE Trans Neural Netw 1:239–242
Li Y, Nara S (2008) Novel tracking function of moving target using chaotic dynamics in a recurrent neural network model. Cogn Neurodyn 2(1):39–48
Osowski S, Bojarczak P, Stodolski M (1996) Fast second order learning algorithm for feedforward multilayer neural network and its applications. Neural Netw 9(9):1583–1596
Shao HM, Zheng GF (2011) Boundedness and convergence of online gradient method with penalty and momentum. Neurocomputing 74:765–770
Sum JP, Leung CS, Ho KI (2012a) Convergence analyses on on-line weight noise injection-based training algorithms for MLPs. IEEE Trans Neural Netw Learn Syst 23(11):1827–1840
Sum JP, Leung CS, Ho KI (2012b) On-line node fault injection training algorithm for MLP networks: objective function and convergence analysis. IEEE Trans Neural Netw Learn Syst 23(2):211–222
Uwate Y, Nishio Y, Ueta T, Kawabe T, Ikeguchi T (2004) Performance of chaos and burst noises injected to the hopfield NN for quadratic assignment problems. IEICE Trans Fundam E87–A(4):937–943
Wang J, Wu W, Zurada JM (2011) Deterministic convergence of conjugate gradient method for feedforward neural networks. Neurocomputing 74:2368–2376
Wu W, Feng G, Li Z, Xu Y (2005) Deterministic convergence of an online gradient method for BP neural networks. IEEE Trans Neural Netw 16:533–540
Wu W, Wang J, Chen MS, Li ZX (2011) Convergence analysis on online gradient method for BP neural networks. Neural Netw 24(1):91–98
Wu Y, Li JJ, Liu SB, Pang JZ, Du MM, Lin P (2013) Noise-induced spatiotemporal patterns in Hodgkin–Huxley neuronal network. Cogn Neurodyn 7(5):431–440
Yoshida H, Kurata S, Li Y, Nara S (2010) Chaotic neural network applied to two-dimensional motion control. Cogn Neurodyn 4(1):69–80
Yu X, Chen QF (2012) Convergence of gradient method with penalty for Ridge Polynomial neural network. Neurocomputing 97:405–409
Zhang NM, Wu W, Zheng GF (2006) Convergence of gradient method with momentum for two-layer feedforward neural networks. IEEE Trans Neural Netw 17(2):522–525
Zhang C, Wu W, Xiong Y (2007) Convergence analysis of batch gradient algorithm for three classes of sigma–pi neural networks. Neural Process Lett 261:77–180
Zhang C, Wu W, Chen XH, Xiong Y (2008) Convergence of BP algorithm for product unit neural networks with exponential weights. Neurocomputing 72:513–520
Zhang HS, Wu W, Liu F, Yao MC (2009) Boundedness and convergence of online gradient method with penalty for feedforward neural networks. IEEE Trans Neural Netw 20(6):1050–1054
Zhang HS, Wu W, Yao MC (2012) Boundedness and convergence of batch back-propagation algorithm with penalty for feedforward neural networks. Neurocomputing 89:141–146
Zhang HS, Liu XD, Xu DP, Zhang Y (2014) Convergence analysis of fully complex backpropagation algorithm based on Wirtinger calculus. Cogn Neurodyn 8(3):261–266
Zheng YH, Wang QY, Danca MF (2014) Noise induced complexity: patterns and collective phenomena in a small-world neuronal network. Cogn Neurodyn 8(2):143–149
Acknowledgments
This work is partly supported by the National Natural Science Foundation of China (Nos. 61101228, 61301202, 61402071), the China Postdoctoral Science Foundation (No. 2012M520623), and the Research Fund for the Doctoral Program of Higher Education of China (No. 20122304120028).
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Zhang, H., Zhang, Y., Xu, D. et al. Deterministic convergence of chaos injection-based gradient method for training feedforward neural networks. Cogn Neurodyn 9, 331–340 (2015). https://doi.org/10.1007/s11571-014-9323-z
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11571-014-9323-z