Abstract
This paper presents a novel design method for multi-valued auto-associative and hetero-associative memories based on a continuous neural network (CNN) with a class of non-smooth linear nondecreasing activation functions. The proposed CNN is robust in terms of the design parameter selection, which is dependent on a set of inequalities rather than the learning procedure. Some globally exponentially stable criteria are obtained to ensure multi-valued associative patterns to be retrieved accurately. The methodology, by generating CNN where the input data are fed via external inputs, avoids spurious memory patterns and achieves \((2r)^n\) storage capacity. These analytic results are applied to the associative memory of images. The fault-tolerant capability and the effectiveness are validated by illustrative experiments.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
References
Zeng Z, Wang J (2008) Design and analysis of high-capacity associative memories based on a class of discrete-time recurrent neural networks. IEEE Trans Syst Man Cybern Part B (Cybernet) 38(6):1525–1536
Segura E, Perazzo R (2000) Associative memories in infinite dimensional spaces. Neural Process Lett 12(2):129–144
Baghelani M, Ebrahimi A, Ghavifekr H (2014) Design of RF MEMS based oscillatory neural network for ultra high speed associative memories. Neural Process Lett 40(1):93–102
Zeng Z, Wang J (2009) Associative memories based on continuous-time cellular neural networks designed using space-invariant cloning templates. Neural Netw 22(5–6):651–657
Grassi G (1997) A new approach to design cellular neural networks for associative memories. IEEE Trans Circuits Syst I Fundam Theory Appl 44(9):835–838
Grassi G (2001) On discrete-time cellular neural networks for associative memories. IEEE Trans Circuits Syst I Fundam Theory Appl 48(1):107–111
Delbem A, Correa L, Zhao L (2009) Design of associative memories using cellular neural networks. Neurocomputing 72:2180–2188
Cao J, Feng G, Wang Y (2008) Multistability and multiperiodicity of delayed Cohen–Grossberg neural networks with a general class of activation functions. Phys D Nonlinear Phenom 237:1734–1749
Nie X, Zheng W, Cao J (2015) Multistability of memristive Cohen–Grossberg neural networks with non-monotonic piecewise linear activation functions and time-varying delays. Neural Netw 71:27–36
Zheng P, Tang W, Zhang J (2010) Efficient continuous-time asymmetric Hopfield networks for memory retrieval. Neural Comput 22:1597–1614
Zheng P (2014) Threshold complex-valued neural associative memory. IEEE Trans Neural Netw Learn Syst 25(9):1714–1718
Bao G, Zeng Z (2012) Analysis and design of associative memory based on recurrent neural network with discontinuous activation functions. Neurocomputing 77:101–107
Akiduki T, Zhong Z, Imamura T, Miyake T (2009) Associative memories with multi-valued cellular neural networks and their application to disease diagnosis. In: IEEE international conference on systems, man and cybernetics, SMC 2009. IEEE, pp 3824–3829 (2009)
Park J, Kim H, Park Y, Lee S (2009) A synthesis procedure for associative memories based on space-varying cellular neural networks. Neural Netw 14:107–113
Zhang Z, Akiduki T, Imamura T, Miyake T (2007) Realization of multi-valued associative memory with cellular neural network. In: IEEE international conference on systems, man and cybernetics, ISIC. IEEE, pp 1205–1210 (2007)
Han Q, Liao X, Huang T, Peng J, Li C, Huang H (2012) Analysis and design of associative memories based on stability of cellular neural networks. Neurocomputing 97:192–200
Zhang H, Huang Y, Wang B, Wang Z (2014) Design and analysis of associative memories based on external inputs of delayed recurrent neural networks. Neurocomputing 136:337–344
Zhou C, Zeng X, Yu J, Jiang H (2016) A unified associative memory model based on external inputs of continuous recurrent neural networks. Neurocomputing 186:44–53
Sha C, Zhao H (2017) Design and analysis of associative memories based on external inputs of continuous bidirectional associative networks. Neurocomputing 266:433–444
Xiu C, Liu C, Cheng Y (2015) Associative memory network and its hardware design. Neurocomputing 158:204–209
Zeidler E (1986) Nonlinear functional analysis and its applications I: fixed-point theorems. Springer, Berlin
LaSalle JP (1976) The stability of dynamical systems. Society for Industrial and Applied Mathematics, Philadelphia
Wang L, Chen T (2012) Multistability of neural networks with Mexican-hat-type activation functions. IEEE Trans Neural Netw Learn Syst 23(11):1816–1826
Xiu C, Liu C, Cheng Y (2015) Associative memory network and its hardware design. Neurocomputing 158:204–209
Acknowledgements
The work was supported by National Natural Science Foundation of China under Grant Nos. 11571170 and 11501290. The authors would like to express our gratitude to Editor and the anonymous referees for their valuable comments and suggestions that led to truly significant improvement of the manuscript.
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Sha, C., Zhao, H., Yuan, Y. et al. Synthesization of Multi-valued Associative High-Capacity Memory Based on Continuous Networks with a Class of Non-smooth Linear Nondecreasing Activation Functions. Neural Process Lett 50, 911–932 (2019). https://doi.org/10.1007/s11063-018-9955-9
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11063-018-9955-9