Supervised learning in spiking neural networks with FORCE training

Abstract

Populations of neurons display an extraordinary diversity in the behaviors they affect and display. Machine learning techniques have recently emerged that allow us to create networks of model neurons that display behaviors of similar complexity. Here we demonstrate the direct applicability of one such technique, the FORCE method, to spiking neural networks. We train these networks to mimic dynamical systems, classify inputs, and store discrete sequences that correspond to the notes of a song. Finally, we use FORCE training to create two biologically motivated model circuits. One is inspired by the zebra finch and successfully reproduces songbird singing. The second network is motivated by the hippocampus and is trained to store and replay a movie scene. FORCE trained networks reproduce behaviors comparable in complexity to their inspired circuits and yield information not easily obtainable with other techniques, such as behavioral responses to pharmacological manipulations and spike timing statistics.

Full text

Its licence does not allow us to show the full text. Read it at PubMed Central:

References 77

1. 1.

   Sussillo D, Abbott LF. Generating coherent patterns of activity from chaotic neural networks. Neuron. 2009;63:544-557.

   Not in our library10.1016/j.neuron.2009.07.018

2. 2.

   DePasquale, B., Churchland, M. M. & Abbott, L. Using firing-rate dynamics to train recurrent networks of spiking model neurons. Preprint at https://arxiv.org/abs/1601.07620 (2016).

   Not in our library

3. 3.

   Abbott L, DePasquale B, Memmesheimer RM. Building functional networks of spiking model neurons. Nat. Neurosci. 2016;19:350-355.

   Not in our library10.1038/nn.4241

4. 4.

   Thalmeier D, Uhlmann M, Kappen HJ, Memmesheimer R-M. Learning universal computations with spikes. PLoS. Comput. Biol. 2016;12:e1004895.

   Not in our library10.1371/journal.pcbi.1004895

5. 5.
   Predictive Coding of Dynamical Variables in Balanced Spiking Networks · 2013

   No notice
6. 6.

   Schwemmer MA, Fairhall AL, Denéve S, Shea-Brown ET. Constructing precisely computing networks with biophysical spiking neurons. J. Neurosci. 2015;35:10112-10134.

   Not in our library10.1523/jneurosci.4951-14.2015

7. 7.

   Bourdoukan, R. & Denéve, S. Enforcing balance allows local supervised learning in spiking recurrent networks. In Advances in Neural Information Processing Systems, 982–990 (2015).

   Not in our library

8. 8.

   Eliasmith C, et al. A large-scale model of the functioning brain. Science. 2012;338:1202-1205.

   Not in our library10.1126/science.1225266

9. 9.

   Eliasmith C, Anderson CH. Neural engineering: Computation, representation, and dynamics in neurobiological systems. 2002.

   Not in our library

10. 10.

    Gilra, A. & Gerstner, W. Predicting non-linear dynamics: a stable local learning scheme for recurrent spiking neural networks. arXiv:1702.06463 (2017).

    Not in our library10.7554/elife.28295

11. 11.

    Alemi, A., Machens, C., Denéve, S. & Slotine, J.-J. Learning arbitrary dynamics in efficient, balanced spiking networks using local plasticity rules. Preprint at https://arxiv.org/abs/1705.08026 (2017).

    Not in our library

12. 12.

    Brendel, W., Bourdoukan, R., Vertechi, P., Machens, C. K. & Denéve, S. Learning to represent signals spike by spike. arXiv:1703.03777 (2017).

    Not in our library10.1371/journal.pcbi.1007692

13. 13.

    Huh, D. & Sejnowski, T. J. Gradient descent for spiking neural networks. arXiv:1706.04698 (2017).

    Not in our library

14. 14.

    Lukoševičius M, Jaeger H, Schrauwen B. Reservoir computing trends. KI-Künstliche Intelligenz. 2012;26:365-371.

    Not in our library10.1007/s13218-012-0204-5

15. 15.

    Lukoševičius M, Jaeger H. Reservoir computing approaches to recurrent neural network training. Comput. Sci. Rev. 2009;3:127-149.

    Not in our library10.1016/j.cosrev.2009.03.005

16. 16.

    Schrauwen, B., Verstraeten, D. & Van Campenhout, J. An overview of reservoir computing: theory, applications and implementations. In Proceedings of the 15th European Symposium on Articial Neural Networks, 471–482 (2007).

    Not in our library

17. 17.

    Dominey PF. Complex sensory-motor sequence learning based on recurrent state representation and reinforcement learning. Biol. Cybern. 1995;73:265-274.

    Not in our library10.1007/bf00201428

18. 18.

    Maass W, Natschläger T, Markram H. Real-time computing without stable states: a new framework for neural computation based on perturbations. Neural. Comput. 2002;14:2531-2560.

    Not in our library10.1162/089976602760407955

19. 19.

    Jaeger H. The “echo state” approach to analysing and training recurrent neural networks-with an erratum note. Bonn, Ger. 2001;148:34.

    Not in our library

20. 20.

    Schliebs, S., Mohemmed, A. & Kasabov, N. Are probabilistic spiking neural networks suitable for reservoir computing? In Neural Networks (IJCNN), The 2011 International Joint Conference on Neural Networks, 3156–3163 (2011).

    Not in our library

21. 21.

    Ozturk MC, Principe JC. Computing with transiently stable states in Proceedings. IEEE International Joint Conference on Neural Networks. 2005;3:1467-1472.

    Not in our library

22. 22.

    Maass, W. Liquid state machines: motivation, theory, and applications. In: B. Cooper, A. Sorbi (eds.) Computability in Context: Computation and Logic in the Real World, 275–296 (Imperial College Press, 2011).

    Not in our library

23. 23.

    Maass W, Natschläger T, Markram H. Fading memory and kernel properties of generic cortical microcircuit models. J. Physiol. 2004;98:315-330.

    Not in our library10.1016/j.jphysparis.2005.09.020

24. 24.

    Wojcik GM, Kaminski WA. Liquid state machine built of Hodgkin–Huxley neurons and pattern recognition. Neurocomputing. 2004;58:245-251.

    Not in our library10.1016/j.neucom.2004.01.051

25. 25.

    Buonomano DV, Merzenich MM. Temporal information transformed into a spatial code by a neural network with realistic properties. Science. 1995;267:1028.

    Not in our library10.1126/science.7863330

26. 26.

    Van Vreeswijk C, Sompolinsky H. Chaos in neuronal networks with balanced excitatory and inhibitory activity. Science. 1996;274:1724.

    Not in our library10.1126/science.274.5293.1724

27. 27.

    Ostojic S. Two types of asynchronous activity in networks of excitatory and inhibitory spiking neurons. Nat. Neurosci. 2014;17:594-600.

    Not in our library10.1038/nn.3658

28. 28.

    Harish O, Hansel D. Asynchronous rate chaos in spiking neuronal circuits. PLoS. Comput. Biol. 2015;11:e1004266.

    Not in our library10.1371/journal.pcbi.1004266

29. 29.

    Haykin, S. S. Neural networks and learning machines (Pearson Upper Saddle River, NJ, 2009).

    Not in our library

30. 30.

    Haykin, S. in Neural Networks and Learning Machines 3rd edn, 246-249 (Upper Saddle River, NJ, USA; Pearson, 2009).

    Not in our library

31. 31.

    Hines ML, Morse T, Migliore M, Carnevale NT, Shepherd GM. ModelDB: a database to support computational neuroscience. J. Comput. Neurosci. 2004;17:7-11.

    Not in our library10.1023/b:jcns.0000023869.22017.2e

32. 32.

    Rajan K, Abbott L. Eigenvalue spectra of random matrices for neural networks. Phys. Rev. Lett. 2006;97:188104.

    Not in our library10.1103/physrevlett.97.188104

33. 33.

    London M, Roth A, Beeren L, Häusser M, Latham PE. Sensitivity to perturbations in vivo implies high noise and suggests rate coding in cortex. Nature. 2010;466:123-127.

    Not in our library10.1038/nature09086

34. 34.

    Monteforte M, Wolf F. Dynamic flux tubes form reservoirs of stability in neuronal circuits. Phys. Rev. X. 2012;2:041007.

    Not in our library

35. 35.

    Hahnloser RH, Kozhevnikov AA, Fee MS. An ultra-sparse code underlies the generation of neural sequences in a songbird. Nature. 2002;419:65-70.

    Not in our library10.1038/nature00974

36. 36.

    Long MA, Jin DZ, Fee MS. Support for a synaptic chain model of neuronal sequence generation. Nature. 2010;468:394-399.

    Not in our library10.1038/nature09514

37. 37.

    Leonardo A, Fee MS. Ensemble coding of vocal control in birdsong. J. Neurosci. 2005;25:652-661.

    Not in our library10.1523/jneurosci.3036-04.2005

38. 38.

    Vicario DS, Raksin JN. Possible roles for GABAergic inhibition in the vocal control system of the zebra finch. Neuroreport. 2000;11:3631-3635.

    Not in our library10.1097/00001756-200011090-00046

39. 39.

    Nottebohm F, Stokes TM, Leonard CM. Central control of song in the canary, Serinus canarius. J. Comp. Neurol. 1976;165:457-486.

    Not in our library10.1002/cne.901650405

40. 40.

    Euston DR, Tatsuno M, McNaughton BL. Fast-forward playback of recent memory sequences in prefrontal cortex during sleep. Science. 2007;318:1147-1150.

    Not in our library10.1126/science.1148979

Cited by papers we hold 6

• Intrinsic bursts facilitate learning of Lévy flight movements in recurrent neural network models · 2022
  Not checked
• Neural heterogeneity promotes robust learning · 2021
  No notice
• Optimal learning with excitatory and inhibitory synapses. · 2020
  Not checked
• A solution to the learning dilemma for recurrent networks of spiking neurons · 2020
  No notice
• Contrasting the effects of adaptation and synaptic filtering on the timescales of dynamics in recurrent networks. · 2019
  Not checked
• How single neuron properties shape chaotic dynamics and signal transmission in random neural networks. · 2019
  Not checked

Reads

A person counts once a day per paper. We keep no cookie and no address: a reader is a hash that changes every day. Any month under five reads shows no number.