PATTERN RECOGNITION ABILITIES OF RECURRENT NEURAL NETWORKS
Abstract
The artificial neural networks are simulating the human brain. Could artificial neural networks memorize as the human brain? The paper presents the structures, the learning rules and the stability of the Hopfield, Bidirectional Associative Memory (BAM), two main recurrent neural networks. We also perform the experiments on their memory abilities, ability of fault isolation for several of failure bits. An example of pattern recognition of image faces and corresponding their labels are also represented.
References
Nguyễn Quang Hoan (2007), Mạng nơron nhân tạo, Học viện Công nghệ Bưu chính Viễn thông.
Chin Teng Lin, C.S. George Lee (2002), Neural Fuzzy Systems, A Neural-Fuzzy Synergism to Intelligent Systems, Prentice Hall International, Inc.
Michael Negnevitsky (2002), Artificial Intelligence. A Guide to Intelligent Systems. Addison Wesley. Pearson Education.
Xiaomei Wang, ShoumingZhong (2010), Globally Exponential Stability of Periodic Solution of BAM Neural Network with Distributed Time Delays and Impulses, 2010 International Conference on Computer and Computational Intelligence (ICCCI 2010).
Lejie Yu (2010), The New Research of Associative Memory Neural Networks, 2010 International Conference on Computer and Computational Intelligence (ICCCI 2010).
Bart Kosko (1988). Bidirectional Associative Memories. IEEE Trans. On System, Man, and Cybernetics, vol. 18, No.1, January/February.
Nathan John Burles (2014). Pattern Recognition Using Associative Memories. PhD. Dissertation, University of York.
Hui Wang, Yue Wu, Biaobiao Zhang and K. -L. Du (2011). Recurrent Neural Networks: Associative Memory and Optimization, J. Inform Tech Soft Eng. 1:104: 10.4172/2165-7866.1000104.
Nguyen Quang Hoan. The Stability of the High-Order Hopfield Neural Networks and Their Application Abilities for Robot. PhD. Dissertation, 1996.
