Affiliation:
1. Kharkiv National University of Radio Electronics, Ukraine
Abstract
The paper proposes a 2D-hybrid system of computational intelligence, which is based on the generalized neo-fuzzy neuron. The system is characterised by high approximate abilities, simple computational implementation, and high learning speed. The characteristic property of the proposed system is that on its input the signal is fed not in the traditional vector form, but in the image-matrix form. Such an approach allows getting rid of additional convolution-pooling layers that are used in deep neural networks as an encoder. The main elements of the proposed system are a fuzzified multidimensional bilinear model, additional softmax layer, and multidimensional generalized neo-fuzzy neuron tuning with cross-entropy criterion. Compared to deep neural systems, the proposed matrix neo-fuzzy system contains gradually fewer tuning parameters – synaptic weights. The usage of the time-optimal algorithm for tuning synaptic weights allows implementing learning in an online mode.
Publisher
Riga Technical University
Reference28 articles.
1. P. Daniušis and P. Vaitkus, “Neural network with matrix inputs,” Informatica, vol. 19, no. 4, pp. 477–486, Jan. 2008. https://doi.org/10.15388/Informatica.2008.225
2. P. Stubberud, “A vector matrix real time backpropagation algorithm for recurrent neural networks that approximate multi-valued periodic functions,” International Journal on Computational Intelligence and Application, vol. 8, no. 4, pp. 395–411, Dec. 2009. https://doi.org/10.1142/S1469026809002667
3. M. Mohamadian, H. Afarideh, and F. Babapour, “New 2D matrix-based neural network for image processing applications,” IAENG International Journal of Computer Science, vol. 42, no. 3, pp. 265–274, 2015.
4. J. Yao, Y. Guo, and Z. Wang, “Matrix neural networks,” Lecture Notes in Computer Science, vol. 10261, May 2017, pp. 313–320. https://doi.org/10.1007/978-3-319-59072-1_37
5. J. Pliss, O. Boiko, V. Volkova, and Y. Bodyanskiy, “Matrix deep neural network and its rapid learning in Data Science tasks",” Proceedings of the International Conference “Advanced Computer Information Technologies,” Czech Republic, June 2018, pp. 141–144.