Fast and Energy Efficient Learning Algorithm for Kohonen Neural Network Realized in Hardware

• Autorzy: Kolasa M.
• Języki publikacji: EN

Abstrakty

EN

A new fast energy efficient learning algorithm suitable for hardware implemented Kohonen Self-Organizing Map (SOM) is proposed in the paper. The new technique is based on a multistage filtering of the quantization error. The algorithm detects such periods in the learning process, in which the quantization error is decreasing (the ‘activity’ phases), which can be interpreted as a progress in training, as well as the ‘stagnation’ phases, in which the error does not decrease. The neighborhood radius is reduced by 1 always just after the training process enters one of the ‘stagnation’ phases, thus shortening this phase. The comprehensive simulations on the software model (in C++) have been carried out to investigate the influence of the proposed algorithm on the learning process. The learning process has been assessed by the used of five criteria, which allow assessing the learning algorithm in two different ways i.e., by expressing the quality of the vector quantization, as well as the topographic mapping. The new algorithm is able to shorten the overall training process by more than 90% thus reducing the energy consumed by the SOM also by 90%. The proposed training algorithm is to be used in a new high performance Neuroprocessor that will find a broad application in a new generation of Wireless Body Area Networks ( WBAN) used in the monitoring of the biomedical signals like, for example, the Electrocardiogram (ECG) signals.

Słowa kluczowe

EN

Kohonen neural network; CMOS implementation; WBAN; optimized learning process; low energy consumption

PL

sieci neuronowe; CMOS; WBAN; zoptymalizowany proces uczenia; niskie zużycie energii

• Wydawca: Oficyna Wydawnicza Politechniki Białostockiej
• Czasopismo: Acta Mechanica et Automatica
• Rocznik: 2012
• Tom: Vol. 6, no. 3
• Strony: 52--57
• Opis fizyczny: Bibliogr. 22 poz., Rys.

Twórcy

autor

Kolasa M.

• Institute of Electrical Engineering, Faculty of Telecommunication and Electrical Engineering, University of Technology and Life Sciences, ul. Kaliskiego 7, 85-796, Bydgoszcz, Poland,

Bibliografia

• 1. Beaton D., Valova I., MacLean D. (2010), CQoCO: A measure for comparative quality of coverage and organization for selforganizing maps, Neurocomputing, Vol. 73, 2147–2159.
• 2. Bolkowski S., Stabrowski M., Skoczylas J., Sroka J., Sikora J., Wincenciak S. (1993), Computer analysis methods of electromagnetic field, WNT, Warsaw.
• 3. Boniecki P. (2005), The Kohonen neural network in classification problems solving in agricultural engineering, Journal of Research and Applications in Agricultural Engineering, Vol. 50, No. 1, 37-40.
• 4. Brocki L. (2007), Recent Advances in Mechatronics, Springer BerlinHeidelberg.
• 5. Chudáček V., Georgoulas G., Lhotská L., Stylios C., Petrík M., Čepek M. (2009), Examining cross-database global training to evaluate five different methods for ventricular beat classification, Physiological Measurem, Vol. 30, No. 7, 661-677.
• 6. Długosz R., Kolasa M., Pedrycz W., Szulc M. (2011), Parallel Programmable Asynchronous Neighborhood Mechanism for Kohonen SOM Implemented in CMOS Technology, IEEE Transactions on Neural Networks, Vol. 22, No. 12, 2091-2104.
• 7. Fernandez E.A., Willshaw P., Perazzo C.A., Presedo R.J. ́ , Barro S. (2001), Detection of abnormality in the electrocardiogram without prior knowledge by using the quantisation error of a self-organising map, tested on the European ischaemia database, Medical and Biological Engineering and Computing, Vol. 39, No. 3, 330-337.
• 8. Kohonen T. (2001), Self-Organizing Maps, third ed. Springer, Berlin.
• 9. Kolasa M., Długosz R., Pedrycz W., Szulc M. (2012), A programmable triangular neighborhood function for a Kohonen self-organizing map implemented on chip, Neural Networks ,Vol. 25, 146-160.
• 10. Lagerholm M., Peterson G. (2000), Clustering ECG complexes using hermite functions and self-organizing maps, IEEE Transactions on Biomedical Engineering, Vol. 47, No. 7, 838-848.
• 11. Lee J. A., Donckers N., Verleysen M., (2001), Recursive learning rules for SOMs, Workshop on Self-Organizing Maps, ser. Advances in Self-Organising Maps, Springer Verlag, 67–72.
• 12. Lee J., Verleysen M. (2002), Self-organizing maps with recursive neighborhood adaptation, Neural Networks, Vol. 15, No. 8-9, 993–1003.
• 13. Leite C.R., Martin D.L., Sizilio G.R., Dos Santos K.E., de Araujo B.G., Valentim R.A., Neto A.D., de Melo J.D., Guerreiro A.M.(2010), Classification of cardiac arrhythmias using competitive networks, 2010 Annual International Conference of the IEEE Engineering in Medicine and Biology Society, 1386-1389.
• 14. Li F., Chang C.-H., Siek L. (2009), A compact current mode neuron circuit with Gaussian taper learning capability, IEEE Int. Symp. Circuits and Systems, 2129–2132.
• 15. Mokriš, Forgáč R. (2004), Decreasing the Feature Space Dimension by Kohonen Self-Orgaizing Maps, 2nd Slovakian – Hungarian Joint Symposium on Applied Machine Intelligence, Herľany, Slovakia.
• 16. Osowski S., Linh T.H. (2001), ECG beat recognition using fuzzy hybrid neural network, IEEE Transactions on Biomedical Engineering, Vol. 48, No. 11, 1265-1271.
• 17. Su M.-C., Chang H.-T., Chou C.-H. (2002), A novel measure for quantifying the topology preservation of self-organizing feature maps, Neural Process. Lett., Vol. 15, No. 2,137–145.
• 18. Talbi M. L., Charef A., Ravier P. (2010), Arrhythmias classification using the fractal behavior of the power spectrum density of the QRS complex and ANN, International Conference on High Performance Computing and Simulation, 399-404.
• 19. Tighiouart B., Rubel P., Bedda M. (2003), Improvement of QRS boundary recognition by means of unsupervised learning, Computers in Cardiology, Vol. 30, 49-52.
• 20. Uriarte E., Martin F. (2005), Topology Preservation in SOM, Int. J. of Math. and Comput. Sci., Vol. 1, No. 1, 19–22.
• 21. Valenza G., Lanata A., Ferro M., Scilingo E.P. (2008), Real-time discrimination of multiple cardiac arrhythmias for wearable systems based on neural networks, Computers in Cardiology, Vol. 35, 1053-1056.
• 22. Wen C., Lin T.-C., Chang K.-C., Huang C.-H. (2009), Classification of ECG complexes using self-organizing CMAC, Measurement: Journal of the International Measurement Confederation, Vol. 42, No. 3, 399-407.