Anomalous and traditional diffusion modelling in SOM learning

• Autorzy: Hrebik Radek , Kukal Jaromir

Identyfikatory

DOI

10.24425/acs.2019.131233

• Języki publikacji: EN

Abstrakty

EN

The traditional self organizing map (SOM) is learned by Kohonen learning. The main disadvantage of this approach is in epoch based learning when the radius and rate of learning are decreasing functions of epoch index. The aim of study is to demonstrate advantages of diffusive learning in single epoch learning and other cases for both traditional and anomalous diffusion models. We also discuss the differences between traditional and anomalous learning in models and in quality of obtained SOM. The anomalous diffusion model leads to less accurate SOM which is in accordance to biological assumptions of normal diffusive processes in living nervous system. But the traditional Kohonen learning has been overperformed by novel diffusive learning approaches.

Słowa kluczowe

EN

self organization; Kohonen map; diffusion learning; anomalous diffusion; SOM

• Wydawca: Polish Academy of Sciences, Committee of Automatic Control and Robotics
• Czasopismo: Archives of Control Sciences
• Rocznik: 2019
• Tom: Vol. 29, No. 4
• Strony: 699--717
• Opis fizyczny: Bibliogr. 40 poz., rys., tab., wzory

Twórcy

autor

Hrebik Radek

• FNSPE, CTU in Prague, Trojanova 13, 120 00, Prague 2, Czech Republic

autor

Kukal Jaromir

• FNSPE, CTU in Prague, Trojanova 13, 120 00, Prague 2, Czech Republic

Bibliografia

• [1] E. Alonso: Computational Neuroscience for Advancing Artificial Intelligence: Models, Methods and Applications: Models, Methods and Applications. Premier reference source. Medical Information Science Reference, 2010.
• [2] D. Brogioli and A. Vailati:. Diffusive mass transfer by nonequilibrium fluctuations: Fick’s law revisited. Phys. Rev. E, 63, 012105 (2000).
• [3] W. L. Chang, L. M. Pang and K. M. Tay: Application of self-organizing map to failure modes and effects analysis methodology. Neurocomputing, 249 (2017), 314–320.
• [4] J. A. F. Costa, A. P. V. Pinto, J. R. de Andrade and M. G. de Medeiros: Clustering of regional hdi data using self-organizing maps. In 2017 IEEE Latin American Conference on Computational Intelligence (LA-CCI), (2017), 1–6.
• [5] J. Crank: The mathematics of diffusion, J. Crank ed. Clarendon Press Oxford [England], 2nd ed. edition, 1975.
• [6] E. Cussler: Diffusion: Mass Transfer in Fluid Systems. Cambridge Series in Chemical Engineering. Cambridge University Press, 2009.
• [7] B. Davies: Integral Transforms and Their Applications. Texts in Applied Mathematics, Springer New York, 2002.
• [8] G. Edelman and J. Gally: Nitric oxide: linking space and time in the brain. Proceedings of the National Academy of Sciences, 89(24), (1992), 11651–11652.
• [9] J. Espenson: Chemical Kinetics and Reaction Mechanisms. Advanced Chemistry Series, McGraw-Hill, 1995.
• [10] R. A. Fisher: The use of multiple measurements in taxonomic problems. Annals of eugenics, 7(2), (1936), 179-188.
• [11] J. A. Gally, P. R. Montague, G. N. Reeke and G. M. Edelman: The no hypothesis: possible effects of a short-lived, rapidly diffusible signal in the development and function of the nervous system. Proceedings of the National Academy of Sciences, 87(9), (1990), 3547-3551.
• [12] J. Garthwaite, S. L. Charles and R. Chess-Williams: Endothelium-derived relaxing factor release on activation of nmda receptors suggests role as intercellular messenger in the brain. Nature, 336(6197), (1988), 385–388.
• [13] A. Gelperin: Nitric oxide mediates network oscillations of olfactory interneurons in a terrestrial mollusc. Nature, 369(6475), (1994), 61–63.
• [14] D. Graupe: Deep Learning Neural Networks: Design and Case Studies, 2016.
• [15] L. Hamel: Som quality measures: An efficient statistical approach. In Proceedings of the 11th International Workshop WSOM 2016, pages 49-59, Houston. Springer, 2016.
• [16] N. A. Hartell: Strong activation of parallel fibers produces localized calcium transients and a form of ltd that spreads to distant synapses. Neuron, 16(3), (1996), 601-610.
• [17] C. Holscher: Nitric oxide, the enigmatic neuronal messenger: its role in synaptic plasticity. Trends in neurosciences, 20(7), (1997), 298-303.
• [18] R. Hrebik and J. Kukal: Diffusion modelling: Topographic error of som under control. Soft Computing (2018), page submitted.
• [19] T. Kohonen: Self-organized formation of topologically correct feature maps. Biological Cybernetics, 43(1), (1982), 59-69.
• [20] T. Kohonen: Self-Organizing Maps. Springer Series in Information Sciences. Springer Berlin Heidelberg, 2012.
• [21] S. Kornblith, R. Q. Quiroga, C. Koch, I. Fried and F. Mormann: Persistent single-neuron activity during working memory in the human medial temporal lobe. Current Biology, 27(7), (2017), 1026-1032.
• [22] S. Kotz and S. Nadarajah: Multivariate T-Distributions and Their Applications. Cambridge University Press, 2004.
• [23] J. R. Lancaster: Simulation of the diffusion and reaction of endogenously produced nitric oxide. Proceedings of the National Academy of Sciences, 91(17), 1994), 8137-8141.
• [24] A. Lavecchia: Machine-learning approaches in drug discovery: methods and applications. Drug Discovery Today, 20(3), (2015), 318-331.
• [25] P. F. Lopez, C. P. S. Araujo, P. G. Baez and G. S. Martin: Diffusion associative network: diffusive hybrid neuromodulation and volume learning. In International Work-Conference on Artificial Neural Networks, pp. 54-61, Springer, 2003.
• [26] P. F. Lopez, P. G. Baez and C. P. S. Araujo: Nitric oxide diffusion and multicompartmental systems: Modeling and implications. In International Conference on Neural Information Processing, pp. 523-531, Springer, 2015.
• [27] O. A. Moldes, J. C. Mejuto, R. Rial-Otero and J. Simal-Gandara: Acritical review on the applications of artificial neural networks in winemaking technology. Critical Reviews in Food Science and Nutrition, 57(13), (2017), 2896-2908.
• [28] E. Oja and S. Kaski: Kohonen Maps. Elsevier Science, 1999.
• [29] M. O’Shea, R. Colbert, L. Williams and S. Dunn: Nitric oxide compartments in the mushroom bodies of the locust brain. Neuroreport, 9(2), (1998), 333–336.
• [30] J. H. Park, V. A. Straub and M. O’Shea: Anterograde signaling by nitric oxide: Characterization and in vitro reconstitution of an identified nitrergic synapse. Journal of Neuroscience, 18(14), (1998), 5463–5476.
• [31] P. Perner: Machine Learning and Data Mining in Pattern Recognition: 11th International Conference, MLDM 2015, Hamburg, Germany, July 20-21, 2015, Proceedings. Lecture Notes in Computer Science, Springer International Publishing, 2015.
• [32] A. Philippides, P. Husbands and M. O’Shea: Four-dimensional neuronal signaling by nitric oxide: a computational analysis. Journal of Neuroscience, 20(3), (2000), 1199-1207.
• [33] C. Pozrikidis: The Fractional Laplacian. CRC Press, 2016.
• [34] G. Polzlbauer: Survey and comparison of quality measures for selforganizing maps.
• [35] M. Senapati: Advanced Engineering Chemistry. Laxmi Publications, 2006.
• [36] S. H. Snyder and D. S. Bredt: Nitric oxide as a neuronal messenger. Trends in Pharmacological Sciences, 12 (1991), 125-128.
• [37] J. J. Thomson: On the structure of the atom: an investigation of the stability and periods of oscillation of a number of corpuscles arranged at equal intervals around the circumference of a circle; with application of the results to the theory of atomic structure. Philosophical Magazine Series 6, 7(39), (1904), 237-265.
• [38] A. Urmos, Z. Farkas, M. Farkas, T. Sandor, L. T. Koczy and A. Nemcsics: Fuzzy and kohonen som based classification of different 0d nanostructures. In 2017 IEEE 15th International Symposium on Applied Machine Intelligence and Informatics (SAMI), pp. 365-370, 2017.
• [39] L. Wood and J. Garthwaite: Models of the diffusional spread of nitric oxide: implications for neural nitric oxide signalling and its pharmacological properties. Neuropharmacology, 33(11), (1994), 1235-1244.
• [40] Y. Yun: The moments of a diffusion process. Statistics & Probability Letters, 138 (2018), 36-41.

Uwagi

EN

1. The authors would like to acknowledge the support of the research grant SGS 17/196/OHK4/3T/14. The second author also acknowledges the support of the OP VVV MEYS funded project CZ.02.1.01/0.0/0.0/16_019/0000765 Research Center for Informatics.

PL

2. Opracowanie rekordu w ramach umowy 509/P-DUN/2018 ze środków MNiSW przeznaczonych na działalność upowszechniającą naukę (2019).