Comparison of supervised learning methods for spike time coding in spiking neural networks

Kasiński, A. ; Ponulak, F.

International Journal of Applied Mathematics and Computer Science, Tome 16 (2006) no. 1, pp. 101-113.

Résumé

In this review we focus our attention on supervised learning methods for spike time coding in Spiking Neural Networks (SNNs). This study is motivated by recent experimental results regarding information coding in biological neural systems, which suggest that precise timing of individual spikes may be essential for efficient computation in the brain. We are concerned with the fundamental question: What paradigms of neural temporal coding can be implemented with the recent learning methods? In order to answer this question, we discuss various approaches to the learning task considered. We shortly describe the particular learning algorithms and report the results of experiments. Finally, we discuss the properties, assumptions and limitations of each method. We complete this review with a comprehensive list of pointers to the literature.

Détail

Keywords: supervised learning, spiking neural networks, time coding, temporal sequences of spikes
Mots-clés : uczenie nadzorowane, sieć neuronowa, kodowanie czasowe, sekwencja czasowa

Bibliographie

[1] Barber D. (2003): Learning in spiking neural assemblies, In:Advances in Neural Information Processing Systems 15, (S.T.S. Becker and K. Obermayer, Eds.). - MIT Press, Cambridge, MA, pp. 149-156.

[2] Belatreche A., Maguire L.P., McGinnity M. and Wu Q.X. (2003): A Method for Supervised Training of Spiking Neural Networks. - Proc. IEEE Conf. Cybernetics Intelligence - Challenges and Advances, CICA'2003, Reading, UK, pp. 39-44.

[3] Bi G.Q. (2002): Spatiotemporal specificity of synaptic plasticity: Cellular rules and mechanisms. - Biol. Cybern., Vol. 87, No. 5-6, pp. 319-332.

[4] Bienenstock E. (1995): A Model of Neocortex. - Network: Computation in Neural Systems, Vol. 6, No. 2, pp. 179-224.

[5] Bohte S. (2003): Spiking Neural Networks. - Ph.D. thesis, University of Amsterdam, Faculty of Mathematics and Natural Sciences, available at: http://homepages.cwi.nl/~bohte.

[6] Bohte S., Kok J. and La Poutré H. (2000): Spike-prop: Errorbackprogation in multi-layer networks of spiking neurons. - Proc. Euro. Symp. Artificial Neural Networks ESANN'2000, Bruges, Belgium, pp. 419-425.

[7] Bohte S., Kok J. and Poutr'e H.L. (2002): Errorbackpropagation in temporally encoded networks of spiking neurons. - Neurocomp., Vol. 48, No. 1-4, pp. 17-37.

[8] Bohte S.M. (2004): The Evidence for Neural Information Processing with Precise Spike-times: A Survey. -Natural Comput., Vol. 3, No. 4, pp. 195-206.

[9] Bonhoeffer T., Staiger V. and Aertsen A. (1989): Synaptic plasticity in rat hippocampal slice cultures: Local Hebbian conjunction of pre- and postsynaptic stimulation leads to distributed synaptic enhancement.-Proc. Nat. Acad. Sci. USA, Vol. 86, No. 20, pp. 8113-8117.

[10] Carnell A. and Richardson D. (2004): Linear algebra for time series of spikes. - Available at: http://www.bath.ac.uk/~Emasdr/spike.ps

[11] Cohen H. (1993): A Course in Computational Algebraic Number Theory. - New York: Springer.

[12] Gabbiani F. and Midtgaard J. (2001): Neural information processing, In: Encyclopedia of Life Sciences, Nature Publishing Group, www.els.net, Vol. 0, pp. 1-12.

[13] Gerstner W. and Kistler W. (2002a): Spiking Neuron Models. Single Neurons, Populations, Plasticity. - Cambridge: Cambridge University Press.

[14] Gerstner W. and KistlerW. (2002b): Mathematical formulations of Hebbian learning. - Biol. Cybern., Vol. 87, No. 5-6, pp. 404-415.

[15] GerstnerW., Kempter R., van Hemmen J. andWagner H. (1996): A neuronal learning rule for sub-millisecond temporal coding. -Nature, Vol. 383, No. 6595, pp. 76-78.

[16] Guetig R., Aharonov R., Rotter S. and Sompolinsky H. (2003): Learning input correlations through non-linear temporally asymmetric Hebbian plasticity. - J. Neurosci., Vol. 23, No. 9, pp. 3697-3714.

[17] Hebb D. (1949): The Organization of Behavior. - Cambridge: Wiley.

[18] Hertz J., Krogh A. and Palmer R. (1991): Introduction to the Theory of Neural Networks. - Redwood-City, CA: Addison-Wesley.

[19] Kasiński A. and Ponulak F. (2005): Experimental demonstration of learning properties of a new supervised learning method for the spiking neural networks, In: Proc. 15-th Int. Conf. Artificial Neural Networks: Biological Inspirations, Vol. 3696, Lecture Notes in Computer Science. - Berlin: Springer, pp. 145-153.

[20] Kepecs A., Van Rossum M., Song S. and Tegner J. (2002): Spike-timing-dependent plasticity: common themes and divergent vistas.-Biol. Cybern., Vol. 87, No. 5-6, pp. 446-458.

[21] KistlerW. (2002): Spike-timing dependent synaptic plasticity: A phenomenological framework. - Biol. Cybern., Vol. 87, No. 5-6, pp. 416-427.

[22] Koerding K. and Koenig P. (2000): Learning with two sites of synaptic integration. - Network: Comput. Neural Syst., Vol. 11, pp. 1-15

[23] Koerding K. and Koenig P. (2001): Supervised and unsupervise learning with two sites of synaptic integration.-J. Comp. Neurosci., Vol. 11, No. 3, pp. 207-215.

[24] Legenstein R., Naeger C. and Maass W. (2005): What can a neuron learn with spike-timing-dependent plasticity? - (submitted).

[25] MaassW. (1997): Networks of spiking neurons: The third generation of neural network models. - Neural Netw., Vol. 10, No. 9, pp. 1659-1671.

[26] Maass W. (1998): On the role of time and space in neural computation, In: Proc. Federated Conf. CLS'98 and MFCS'98, Mathematical Foundations of Computer Science 1998, Vol. 1450, Lecture Notes in Computer Science. - Berlin: Springer, pp. 72-83.

[27] Maass W. (1999): Paradigms for computing with spiking neurons, In: Models of Neural Networks, (L. van Hemmen, Ed.).-Berlin: Springer.

[28] Maass W. (2002): Paradigms for computing with spiking neurons, In: Models of Neural Networks. Early Vision and Attention, (J.L. van Hemmen, J.D. Cowan and E. Domany, Eds.).-New York: Springer, pp. 373-402.

[29] Maass W. (2003): Computation with spiking neurons, In: The Handbook of Brain Theory and Neural Networks, 2nd Ed., (M. Arbib, Ed.). - MIT Press, Cambridge, pp. 1080- 1083.

[30] MaassW. and Bishop C. (Eds.) (1999): Pulsed Neural Networks. -Cambridge: MIT Press.

[31] Maass W. and Zador A. (1999): Computing and learning with dynamic synapses. - NeuroCOLT2 Technical Report Series NC2-TR-1999-041.

[32] Maass W., Natschlaeger T. and Markram H. (2002): Realtime computing without stable states: A new framework for neural computation based on perturbations. - Neural Comput., Vol. 14, No. 11, pp. 2531-2560.

[33] Markram H., Wang Y. and Tsodyks M. (1998): Differential signaling via the same axon of neocortical pyramidal neurons. -Proc. Nat. Acad. Sci., Vol. 95, No. 9, pp. 5323-5328.

[34] Markram H., Luebke J. and Frotscher B.S.M. (1997): Regulation of synaptic efficacy by coincidence of postsynaptic APs and EPSPs.-Science, Vol. 275, No. 5297, pp. 213-215.

[35] Mehta M.R., Lee A.K. and Wilson M.A. (2002): Role of experience of oscillations in transforming a rate code into a temporal code. - Nature, Vol. 417, No. 6891, pp. 741- 746.

[36] Moore S.C. (2002): Back-Propagation in Spiking Neural Networks. - M.Sc. thesis, University of Bath, available at: http://www.simonchristianmoore.co.uk.

[37] Natschlaeger T., Maass W. and Markram H. (2002): The "liquid computer", a novel strategy for real-time computing on time series.-Special issue on Foundations of Information Processing of TELEMATIK, Vol. 8, No. 1, pp. 32-36.

[38] Pavlidis N.G., Tasoulis D.K., Plagianakos V.P., Nikiforidis G. and Vrahatis M.N. (2005): Spiking neural network training using evolutionary algorithms.-Proc. Int. Joint Conf. Neural Networks, IJCNN'05, Montreal, Canada.

[39] Pfister J.P., Barber D. and Gerstner W. (2003): Optimal Hebbian Learning: A Probabilistic Point of View, In: ICANN/ICONIP 2003, Vol. 2714, Lecture Notes in Computer Science.- Berlin: Springer, pp. 92-98.

[40] Pfister J.-P., Toyoizumi T., Barber D. and Gerstner W. (2005): Optimal spike-timing dependent plasticity for precise action potential firing. - (submitted), available at: http://diwww.epfl.ch/~jpfister/papers/Pfister_05a.pdf.

[41] Ponulak F. (2005): ReSuMe-new supervised learning method for spiking neural networks. - Tech.Rep., Institute of Control and Information Engineering, Poznan University of Technology, available at: http://d1.cie.put.poznan.pl/~fp/.

[42] Ponulak F. and Kasiński A. (2005): A novel approach towards movement control with spiking neural networks. - Proc. 3-rd Int. Symp. Adaptive Motion in Animals and Machines, Ilmenau, (Abstract).

[43] Popović D. and Sinkjaer T. (2000): Control of Movement for the Physically Disabled. - London: Springer.

[44] Ruf B. (1998): Computing and Learning with Spiking Neurons - Theory and Simulations. -Ph.D. thesis, Institute for Theoretical Computer Science, Technische Universitaet Graz, Austria.

[45] Ruf B. and Schmitt M. (1997): Learning temporally encoded patterns in networks of spiking neurons. -Neural Proces. Lett., Vol. 5, No. 1, pp. 9-18.

[46] Rumelhart D., Hinton G. and Williams R. (1986): Learning representations by back-propagating errors. - Nature, Vol. 323, pp. 533-536.

[47] Schrauwen B. and Van Campenhout J. (2004): Improving Spike- Prop: Enhancements to an Error-Backpropagation Rule for Spiking Neural Networks. - Proc. 15-th ProRISC Workshop, Veldhoven, the Netherlands.

[48] Seung S. (2003): Learning in spiking neural networks by reinforcement of stochastic synaptic transmission. - Neuron, Vol. 40, No. 6, pp. 1063-1073.

[49] Sougne J.P. (2001): A learning algorithm for synfire chains, In: Connectionist Models of Learning, Development and Evolution, (R.M. French and J.P. Sougne, Eds.). - London: Springer, pp. 23-32.

[50] SpearsW.M., Jong K.A.D., Baeck T., Fogel D.B. and de Garis H. (1993): An overview of evolutionary computation.-Proc. Europ. Conf. Machine Learning, Vienna, Austria, Vol. 667, pp. 442-459.

[51] Tao H.-Z.W., Zhang L.I., Bi G.-Q. and Poo M.-M. (2000): Selective presynaptic propagation of long-term potentiation in defined neural networks. -J. Neurosci., Vol. 20, No. 9, pp. 3233-3243.

[52] Thorpe S. J., Delorme A. and VanRullen R. (2001): Spike-based strategies for rapid processing. - Neural Netw., Vol. 14, No. 6-7, pp. 715-726.

[53] Tiňo P. and Mills A.J. (2005): Learning beyond finite memory in recurrent networks of spiking neurons, In: Advances in Natural Computation - ICNC 2005, (L. Wang, K. Chen and Y. Ong, Eds.), Lecture Notes in Computer Science.- Berlin: Springer, pp. 666-675.

[54] Van Rullen R., Guyonneau R. and Thorpe S.J. (2005): Spike times make sense. - TRENDS in Neurosci., Vol. 28, No. 1, pp. 1-4.

[55] Weisstein E.W. (2006): Gram-Schmidt Orthonormalization, from MathWorld-A Wolfram Web Resource. - Available at: http://mathworld.wolfram.com/Gram-SchmidtOrthonormalization.html.

[56] Xie X. and Seung S. (2004): Learning in neural networks by reinforcement of irregular spiking. - Phys. Rev., Vol. 69, No. 4, pp. 1-10.

[57] Xin J. and Embrechts M.J. (2001): Supervised Learning with Spiking Neuron Networks. - Proc. IEEE Int. Joint Conf. Neural Networks, IJCNN'01, Washington D.C., pp. 1772-1777.