Information processing and learning in spiking neural networks

• Autorzy: Ponulak F. , Kasiński A.

Warianty tytułu

PL

Mechanizmy przetwarzania informacji i uczenia w impulsowych sieciach neuronowych

• Języki publikacji: EN

Abstrakty

EN

The concept that neural information is encoded in the firing rate of neurons has been the dominant paradigm in neurobiology for many years. This paradigm has also heen adopted by the theory of artificial neural networks. Recent physiological experiments demonstrate, however, that in many systems, neural code is founded on the timing of individual action potentials. The finding has given rise to the emergence of a new class of neural models, called spiking neural networks. In this paper we summarize basic properties of spiking neurons. We focus, in particular, on various models for information coding, synaptic plasticity and learning in spiking networks. Finally, we discuss some real-life applications of spiking models.

PL

Jednym z podstawowych paradygmatów obowiązująych przez wiele lat w neurobiologii była koncepcja kodowania informacji za pomocą średniej częstotliwości impulsów nerwowych. Koncepcja ta została zaadoptowana także w teorii sztucznych sieci neuronowych. Aktualne badania w zakresie neurofizjologii wskazują jednak na istotną rolę indywidualnych impulsów nerwowych w kodowaniu informacji. Odkrycie to dało początek nowej klasie sztucznych sieci neuronowych - tak zwanym sieciom impulsowym. W artykule przedstawione są podstawowe właściwości neuronów impulsowych. Szczególna uwaga poświęcona jest mechanizmom przetwarzania informacji oraz modelom plastyczności synaptycznej i uczenia w sieciach impulsowych. Artykuł zakończony jest dyskusją na temat wybranych zastosowań sieci impulsowych w zadaniach inżynierskich oraz w neuromodelowaniu.

Słowa kluczowe

EN

artificial neural networks; ANN; spiking neural networks; SNN; information processing; synaptic plasticity; neurobiology

PL

sztuczne sieci neuronowe; impulsowe sieci neuronowe; przetwarzanie informacji; plastyczność synaptyczna; neurobiologia

• Wydawca: Oficyna Wydawnicza Politechniki Warszawskiej
• Czasopismo: Prace Naukowe Politechniki Warszawskiej. Elektronika
• Rocznik: 2010
• Tom: z. 175, t. 1
• Strony: 23--40
• Opis fizyczny: Bibliogr. 81 poz., rys., wykr.

Twórcy

autor

Ponulak F.

autor

Kasiński A.

• Poznan University of Technology, Institute of Control and Information Engineering, ul. Piotrowo 3a, 60-965 Poznań, Poland,

Bibliografia

• [1] E. D. Adrian, Y. Zotterman. The impulses produced by sensory nerve-endings: Part II. The response of a Single End-Organ. The Journal of Physiology, April, 1926, Vol. 61, No. 2, pp. 151-171.
• [2] D. Baras, R. Meir. Reinforcement learning, spike-time-dependent plasticity, and the BCM rule. Neural Computation, 2007, Vol. 19, pp. 2245-2279.
• [3] R. Barea et al. E.O.G. guidance of a wheelchair using spiking neural networks. In: Proceedings of the European Symposium on Artificial Neural Networks ES ANN'2000. Proceedings Red. M. Verleysen, Brugges, Belgium, D-Facto, 2000, pp. 233-238.
• [4] A. Belatreche et al. A Method for Supervised Training of Spiking Neural Networks. In: Proceedings of IEEE Cybernetics Intelligence - Challenges and Advances (CICA) 2003. Proceedings. Reading, UK, 2003, pp. 39-44.
• [5] D. Belter, F. Ponulak, S. Rotter. Adaptive Movement Control with Spiking Neural Networks, Part II: Composite Control. In: Proceedings of Recent Advances in Neuro-Robotics, Symposium: Sensorimotor Control. Proceedings, Freiburg, Germany, 2008, pp. 14.
• [6] G.-Q. Bi, M.-M. Poo. Synaptic Modifications in Cultured Hippocampal Neurons: Dependence on Spike Timing, Synaptic Strength, and Postsynaptic Cell Type. The Journal of Neuroscience, 1998, Vol. 18, No. 24, pp. 10464-10472.
• [7] S. M. Bohte. The Evidence for Neural Information Processing with Precise Spike-times: A Survey. Natural Computing, Special issue on selected papers from the FUNN 2003 workshop, June, 2004, Vol. 4, No. 3, pp. 195-206.
• [8] S. Bohte. Spiking Neural Networks. PhD thesis, University of Amsterdam, Faculty of Mathematics and Natural Sciences, 2003. Available at http://homepages.cwi.nl/~bohte.
• [9] S. Bohte, J. Kok, H. La Poutré. Spike-prop: error-backprogation in multi-layer networks of spiking neurons. In: Proceedings of the European Symposium on Artificial Neural Networks ESANN'2000. Proceedings. Red. M. Verleysen. D-Facto, 2000, pp. 419-425.
• [10] S. Bohte, J. Kok, H. L. Poutré. Error-backpropagation in Temporally Encoded Networks of Spiking Neurons. Neurocomputing, 2002, Vol. 48, pp. 17-37.
• [11] O. Booij, H. T. Nguyen. A gradient descent rule for spiking neurons emitting multiple spikes. Information Processing Letters, 2005, Vol. 95, No. 6, pp. 552-558.
• [12] A. Borst, F. E.Theunissen. Information theory and neural coding. Nature Neuroscience, 1999, Vol. 2, No. 11, pp. 947-957.
• [13] H. Burgsteiner. Training networks of biological realistic spiking neurons for real-time robot control. In: Proceedings of the 9th International Conference ESANN'2005. Proceedings, 2005.
• [14] M. R. Carey, J. F. Medina, S. G. Lisberger. Instructive signals for motor learning from visual cortical area MT. Nature Neuroscience, 2005, Vol. 8, pp. 813-819.
• [15] A. Citri, R. C. Malenka. Synaptic plasticity: multiple forms, functions, and mechanisms. Neuropsychopharmacology, 2008, Vol. 33, No. 1, pp. 18-41.
• [16] R. C. de Charms. Information coding in the cortex by independent or coordinated populations. Proceedings of the National Academy of Sciences of USA, 1998, Vol. 95, No. 26, pp. 15166-15168.
• [17] M. Farries, A. Fairhall. Reinforcement Learning With Modulated Spike Timing-Dependent Synaptic Plasticity. Journal of Neurophysiology, 2007, Vol. 98, pp. 3648-3665.
• [18] R. Florian. Reinforcement learning through modulation of spike-timing-dependent synaptic plasticity. Neural Computation, 2007, Vol. 19, No. 6, pp. 1468-1502.
• [19] F. Gabbiani, J. Midtgaard. Neural information processing. Encyclopedia of Life Sciences, Nature Publishing Group, www.els.net, 2001, Vol. , pp. 1-12.
• [20] W. Gerstner, W. Kistler. Mathematical formulations of Hebbian learning. Biological Cybernetics, 2002, Vol. 87, pp. 404-415.
• [21] W. Gerstner, W. Kistler. Spiking Neuron Models. Single Neurons, Populations, Plasticity. Cambridge, Cambridge University Press 2002.
• [22] S. Ghosh-Dastidar, H. Adeli. A new supervised learning algorithm for multiple spiking neural networks with application in epilepsy and seizure detection. Neural Netw., 2009, Vol. 22, No. 10, pp. 1419-1431.
• [23] C. Glackin et al. Implementing fuzzy reasoning on a spiking neural network. In: Proceedings of the 18th International Conference on Artificial Neural Networks, ICANN'2005. Proceedings, 2008. Vol. 5164 series Lecture Notes in Computer Science, pp. 258-267.
• [24] R. Gütig, H. Sompolinsky. The tempotron: a neuron that learns spike timing-based decisions. Nature Neuroscience, 2006, Vol. 9, pp. 420-428.
• [25] R. Gütig, H. Sompolinsky. Neural mechanisms of speech processing: time warp invariants in adaptive integration time. In: Proceedings of the 5th Computational and Systems Neuroscience Meeting (Cosyne 2008). Proceedings, Salt Lake City, USA, 2008, pp. 337.
• [26] D. O. Hebb. The organization of behavior. New York, Wiley 1949.
• [27] M. Hines et al. ModelDB: A Database to Support Computational Neuroscience. Journal of Computational Neuroscience, 2004, Vol. 17, No. 1, pp. 7-11; database available from http://senselab.med.yale.edu/ModelDb.
• [28] J. J. Hopfield. Neurodynamics of mental exploration. Proceedings of the National Academy of Sciences of USA, 2010, Vol. 107, No. 4, pp. 1648-1653.
• [29] M. Ito. Mechanisms of motor learning in the cerebellum. Brain Research, 2000, Vol. 886, pp. 237-245.
• [30] E. M. Izhikevich. Solving the Distal Reward Problem through Linkage of STDP and Dopamine Signaling. Cerebral Cortex, 2007, Vol. 17, No. 10, pp. 2443-2452.
• [31] E. M. Izhikevich et al. Bursts as a unit of neural information: selective communication via resonance. Trends in Neuroscience, 2003, Vol. 26, pp. 161-167.
• [32] P. Joshi, W. Maass. Movement generation with circuits of spiking neurons. Neural Computation, 2005, Vol. 17, No. 8, pp. 1715-1738.
• [33] A. Kasiński, F. Ponulak. Comparison of Supervised Learning Methods for Spike Time Coding in Spiking Neural Networks. International Journal of Applied Mathematics and Computer Science, 2006, Vol. 16, No. 1, pp. 101-113.
• [34] A. Knoblauch. Synchronization and pattern separation in spiking associative memories and visual cortical areas. PhD thesis, Abteilung Neuroinformatik, Universitaet Ulm, 2003.
• [35] E. I. Knudsen. Supervised Learning in the Brain. The Journal of Neuroscience, 1994, Vol. 14, No. 7, pp. 3985-3997.
• [36] C. Koch. Biophysics of Computation: Information Processing in Single Neurons. New York, Oxford University Press 1999.
• [37] K. Lee, D.-S. Kwon. Synaptic plasticity model of a spiking neural network for reinforcement learning. Neurocomputing, 2008, Vol. 71, pp. 3037-3043.
• [38] R. Legenstein, C. Naeger, W. Maass. What can a Neuron Learn with Spike-Timing-Dependent Plasticity? Neural Computation, 2005, Vol. 17, pp. 2337-2382.
• [39] R. Legenstein, D. Pecevski, W. Maass. A learning theory for reward-modulated spike-timing-dependent plasticity with application to biofeedback. PLoS Computational Biology, 2008, Vol. 4, No. 10, pp. 1-27.
• [40] R. C. Liu et al. Variability and Information in a Neural Code of the Cat Lateral Geniculate Nucleus. Journal of Neurophysiology, 2001, Vol. 86, pp. 2789-2806.
• [41] W. Maass. Networks of spiking neurons: The third generation ofneural network models. Neural Networks, 1997, Vol. 10(9), pp. 1659-1671.
• [42] W. Maass. Paradigms for computing with spiking neurons. In: Models of Neural Networks. Early Vision and Attention. Red. J. L. van Hemmen, J. D. Cowan, E. Domany, Vol. 4, chapter 9, pp. 373-402. Springer (New York) 2002.
• [43] W. Maass, T. Natschlaeger, H. Markram. Computational models for generic cortical microcircuits. In: Computational Neuroscience: A Comprehensive Approach. Red. J. Feng, chapter 18, pp. 575-605. Chapman and Hall/CRC, Boca Raton 2004.
• [44] H. Markram et al. Regulation of synaptic efiicacy by coincidence of postsynaptic APs and EPSPs. Science, 1997, Vol. 275, pp. 213-215.
• [45] D. Martinez, E. Hugues. A spiking neural network model of the locust antennal lobe : towards neuromorphic electronic noses inspired from insect olfaction. In: Electronic Noses/Sensors for Detection of Explosives. Red. J. Gardner, J. Yinon, pp. 209-234. Kluwer Academic Publishers 2004. Chapter 14.
• [46] J. Montgomery, G. Carton, D. Bodznick. Error-driven motor learning in fish. Biological Bulletin, 2002, Vol. 203, No. 2, pp. 238-239.
• [47] S. C. Moore. Back-Propagation in Spiking Neural Networks. MSc thesis, University of Bath, 2002. Available from: http://www.simonchristianmoore.co.uk
• [48] R. Muresan. Complex object recognition using a biologically plausible neural model. In: Proceedings of the 2nd WSEAS International Conference on Robotics, Distance Learning and Intelligent Communication Systems. Proceedings, Skiathos, Greece, 2002.
• [49] T. Natschlaeger, B. Ruf. Spatial and temporal pattern analysis via spiking neurons. Network: Comp. Neural Systems, 1998, Vol. 9, No. 3, pp. 319-332.
• [50] Neural Prostheses for Restoration of Sensory and Motor Function. Red. J. K. Chapin, K. A. Moxon. CRC Press 2000.
• [51] D. Newman et al. UCI Repository of Machine Learning Databases, available at: http://www.ics.uci.edu/~mlearn/MLRepository.html
• [52] E. A. D. Paolo. Evolving spike-timing-dependent plasticity for single-trial learning in robots. Phil. Trans. R. Soc. Lond. A, 2003, Vol. 361, pp. 2299-2319.
• [53] E. A. D. Paolo. Spike-timing dependent plasticity for evolved robots. Adapt. Behav., 2003, Vol. 10, pp. 243-263.
• [54] J.-P. Pfister et al. Optimal Spike-Timing Dependent Plasticity for Precise Action Potential Firing. Neural Computation, 2006, Vol. 18, pp. 1318-1348.
• [55] F. Ponulak. ReSuMe - new supervised learning method for Spiking Neural Networks. Technical Report, Institute of Control and Information Engineering, Poznań University of Technology, 2005. Available from: http://dl.cie.put.poznan.pl/~fp/.
• [56] F. Ponulak, D. Belter, A. Kasiński. Adaptive Central Pattern Generator based on Spiking Neural Networks. In: Proceedings of EPFL LATSIS Symposium 2006, Dynamical principles for neuroscience and intelligent biomimetic devices. Proceedings, 2006, pp. 121 -122.
• [57] F. Ponulak, D. Belter, S. Rotler. Adaptive Movement Control with Spiking Neural Networks, Part I: Feedforward Control. In: Proceedings of Recent Advances in Neuro-Robotics, Symposium: Sensorimotor Control. Proceedings, Freiburg, Germany, 2008, pp. 47.
• [58] F. Ponulak, A. Kasiński. Supervised Learning in Spiking Neural Networks with ReSuMe: Sequence Learning, Classification and Spike-Shifting. Neural Computation, 2010, Vol. 22, No. 2, pp. 467-510.
• [59] F. Ponulak, A. Kasiński. Generalization Properties of SNN Trained with ReSuMe. In: Proceedings of European Symposium on Artificial Neural Networks, ESANN'2006. Proceedings, 2006, pp. 623-629.
• [60] F. Ponulak, A. Kasiński. ReSuMe learning method for Spiking Neural Networks dedicated to neuroprostheses control. In: Proceedings of EPFL LATSIS Symposium 2006, Dynamical principles for neuroscience and intelligent biomimetic devices. Proceedings, 2006, pp. 119-120.
• [61] F. Ponulak, S. Rotter. Biologically inspired spiking neural model for motor control and motor learning (abstract). In: Proceedings of NIN Conference on Perceptual Learning, Motor Learning and Automaticity. Proceedings, Amsterdam, 2008, pp. 56.
• [62] Principles of neural sciences. Red. E. R. Kandel, T. M. J. Schwartz, T. M. Jessel. New York, Elsevier 1991.
• [63] F. Rieke et al. Spikes: Exploring the neural code. MIT Press 1997.
• [64] F. Rosenblatt. The Perceptron: A Probabilistic Model for Information Storage and Organization in the Brain. Psychological Review, 1958, Vol. 65, No. 6, pp. 386-408.
• [65] B. Ruf. Computing and Learning with Spiking Neurons - Theory and Simulations. PhD thesis, Institute for Theoretical Computer Science, Technische Universitaet Graz, Austria, 1998.
• [66] B. Ruf, M. Schmitt. Self-organization of spiking neurons using action potential timing. IEEE Trans. Neural Networks, 1998, Vol. 9, No. 3, pp. 575-578.
• [67] D. Rumelhart, G. Hinton, R. Williams. Learning representations by back-propagating errors. Nature, 1986, Vol. 323, pp. 533-536.
• [68] B. Schrauwen, J. V. Campenhout. Improving SpikeProp: Enhancements to an Error-Backpropagation Rule for Spiking Neural Networks. In: Proceedings of the 15th ProRISC Workshop. Proceedings, 11, 2004.
• [69] B. Schrauwen, J. V. Campenhout. Backpropagation for Population-Temporal Coded Spiking Neural Networks. In: Proceedings of the International Conference on Neural Networks. Proceedings, 2006, pp. 1797-1804.
• [70] W. Schultz. Getting Formal with Dopamine and Reward. Neuron, 2002, Vol. 36, pp. 241-263.
• [71] A. Soltani, X.-J. Wang. Synaptic computation underlying probabilistic inference. Nature Neuroscience, 2010, Vol. 13, pp. 112-119.
• [72] J. P. Sougne. A learning algorithm for synfire chains. In: Connectionist Models of Learning, Development and Evolution. Red. R. M. French, J. P. Sougne, pp. 23-32. Springer Verlag (London) 2001.
• [73] W. T. Thach. On the specific role of the cerebellum in motor learning and cognition: Clues from PET activation and lesion studies in man. Behavioral Brain Science, 1996, Vol. 19, pp. 411-431.
• [74] P. Tiňo, A. J. Mills. Learning Beyond Finite Memory in Recurrent Networks of Spiking Neurons. In: Advances in Natural Computation, ICNC'2005. Proceedings. Red. L. Wang, K. Chen, Y. Ong. Springer-Verlag, 2005, Lecture Notes in Computer Science, pp. 666-675.
• [75] R. VanRullen, R. Guyonneau, S. J. Thorpe. Spike times make sense. TRENDS in Neurosciences, 2005, Vol. 28, No. 1, pp. 1-4.
• [76] E. Vasilaki et al. Spike-based reinforcement learning in continuous state and action space: When policy gradient methods fail. PLoS Comput Biol, 2009, Vol. 5, No. 12.
• [77] D. Verstraeten et al. Isolated word recognition with the liquid stale machine: A case study. Information Processing Letters, 2005, Vol. 95, No. 6, pp. 521-528.
• [78] P. Werbos. Beyond Regression: New Tools for Prediction and Analysis. PhD thesis, Harvard University (unpublished doctoral dissertation), 1974.
• [79] B. Widrow. Generalization and Information Storage in Networks of Adaline 'Neurons'. In: Self-Organizing Systems. Red. M. Yovitz, G. Jacobi, G. Goldstein, pp. 435-461. Spartan Books 1962.
• [80] B. Widrow, M. E. Hoff. Adaptive Switching Circuits. IRE WESCON Convention Record, 1960, Vol. 4, pp. 96-104.
• [81] J. Xin, M. J. Embrechts. Supervised Learning with Spiking Neuron Networks. In: Proceedings IEEE International Joint Conference on Neural Networks, IJCNN '01. Proceedings, Washington D.C., July 15-19, 2001, pp. 1772-1777.