Journal of Computational Neuroscience

, Volume 11, Issue 3, pp 207–215 | Cite as

Supervised and Unsupervised Learning with Two Sites of Synaptic Integration

  • Konrad P. Körding
  • Peter König


Many learning rules for neural networks derive from abstract objective functions. The weights in those networks are typically optimized utilizing gradient ascent on the objective function. In those networks each neuron needs to store two variables. One variable, called activity, contains the bottom-up sensory-fugal information involved in the core signal processing. The other variable typically describes the derivative of the objective function with respect to the cell's activity and is exclusively used for learning. This variable allows the objective function's derivative to be calculated with respect to each weight and thus the weight update. Although this approach is widely used, the mapping of such two variables onto physiology is unclear, and these learning algorithms are often considered biologically unrealistic. However, recent research on the properties of cortical pyramidal neurons shows that these cells have at least two sites of synaptic integration, the basal and the apical dendrite, and are thus appropriately described by at least two variables. Here we discuss whether these results could constitute a physiological basis for the described abstract learning rules. As examples we demonstrate an implementation of the backpropagation of error algorithm and a specific self-supervised learning algorithm using these principles. Thus, compared to standard, one-integration-site neurons, it is possible to incorporate interesting properties in neural networks that are inspired by physiology with a modest increase of complexity.

calcium spike apical dendrite backpropagation self-supervised learning bursts of action potentials 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. Agmon-Snir H, Carr CE, Rinzel J (1998) The role of dendrites in auditory coincidence detection. Nature 393: 268-272.Google Scholar
  2. Amitai Y, Friedman A, Connors BW, Gutnick MJ (1993) Regenerative activity in apical dendrites of pyramidal cells in neocortex. Cereb. Cortex 3: 26-38.Google Scholar
  3. Arbib MA (1998) The Handbook of Brain Theory and Neural Networks. MIT Press: Cambridge, MA.Google Scholar
  4. Artola A, Bröcher S, Singer W (1990) Different voltage-dependent thresholds for inducing long-term depression and long-term potentiation in slices of rat visual cortex. Nature 347: 69-72.Google Scholar
  5. Bartlett MS, Sejnowski TJ (1998) Viewpoint invariant face representation from visual experience by temporal association. In: Wechsler et al., eds. Face Recognition: From Theory to Applications. Springer, Berlin.Google Scholar
  6. Becker S (1996) Models of cortical self-organisation. Network: Comput. Neural Syst. 7: 7-31.Google Scholar
  7. Becker S, Hinton GE (1992) Self-organizing neural network that discovers surfaces in random-dot stereograms. Nature 355: 161-163.Google Scholar
  8. Berger T, Larkum ME, Lüscher H-R (2001) A high Ih channel density in the distal apical dendrite of layer 5 neocortical pyramidal cells increases bidirectional attenuation of EPSPs. J. Neurophysiol. 85: 855-868.Google Scholar
  9. Buzsaki G, Kandel A (1998) Somadendritic backpropagation of action potentials in cortical pyramidal cells of the awake rat. J. Neurophysiol. 79: 1587-1591.Google Scholar
  10. Carnevale NT, Tsai KY, Claiborne BJ, Brown TH (1997) Comparative electrotonic analysis of three classes of rat hippocampal neurons. J. Neurophysiol. 78: 703-720.Google Scholar
  11. Cash S, Yuste R (1998) Input summation by cultured pyramidal neurons is linear and position-independent. J. Neurosci. 18: 10-15.Google Scholar
  12. Cauller LJ, Connors BW (1994) Synaptic physiology of horizontal afferents to layer I in slices of rat SI neocortex. J. Neurosci. 14: 751-762.Google Scholar
  13. Cook EP, Johnston D (1999) Voltage-dependent properties of dendrites that eliminate location-dependent variability of synaptic input. J. Neurophysiol. 81: 535-543.Google Scholar
  14. Crick FHC (1989) The recent excitement about neural networks. Nature 337: 129-132.Google Scholar
  15. De Sa VR, Ballard DH (1998) Category learning through multimodality sensing. Neur. Comput. 10: 1097-1117.Google Scholar
  16. Dudek SM, Bear MF (1993) Bidirectional long-term modification of synaptic effectiveness in the adult and immature hippocampus. J. Neurosci. 13: 2910-2918.Google Scholar
  17. Földiak P (1991) Learning invariance from transformation sequences. Neur. Comput. 3: 194-200.Google Scholar
  18. Frey, B. (1998) Graphical Models for Machine Learning and Digital Communication. MIT Press: Cambridge, MA.Google Scholar
  19. Gupta A, Wang Y, Markram H (2000) Organizing principles for a diversity of GABAergic interneurons and synapses in the neocortex. Science 287: 273-278.Google Scholar
  20. Helmchen F, Svoboda K, Denk W, Tank DW (1999) In vivo dendritic calcium dynamics in deep-layer cortical pyramidal neurons. Nat. Neurosci. 2: 989-996.Google Scholar
  21. Hinton GE, McClelland JL (1988) Learning representations by recirculation. In: Anderson DZ, ed. Neural Information Processing Systems. American Institute of Physics, New York. pp. 358-366.Google Scholar
  22. Hirsch JA, Alonso JM, Reid RC (1995) Visually evoked calcium action potentials in cat striate cortex. Nature 378: 612-616.Google Scholar
  23. Johnston D, Magee JC, Colbert CM, Cristie BR (1996) Active properties of neuronal dendrites. Ann. Rev. Neurosci. 19: 165-186.Google Scholar
  24. Kay J, Floreano D, Phillips WA (1998) Contextually guided unsupervised learning using local multivariate binary processors. Neural Networks 11: 117-140.Google Scholar
  25. König P, Engel AK (1995) Correlated firing in sensory-motor systems. Curr. Opin. Neurobiol. 5: 511-519.Google Scholar
  26. Körding KP, König P (2000) Learning with two sites of synaptic integration. Network: Comput. Neural Syst. 11: 1-15.Google Scholar
  27. Larkum ME, Kaiser KM, Sakmann B (1999a) Calcium electrogenesis in distal apical dendrites of layer 5 pyramidal cells at a critical frequency of backpropagating action potentials. Proc. Natl. Acad. Sci. USA 96: 14600-14604.Google Scholar
  28. Larkum ME, Zhu JJ, Sakmann B (1999b) A new cellular mechanism for coupling inputs arriving at different cortical layers. Nature 398: 338-341.Google Scholar
  29. LeCun Y, Boser B, Denker JS, Henderson D, Howard RE, Hubbard W, Jackel LD (1990) Handwritten digit recognition with a backpropagation network. In: Touretzky D, ed. Advances in Neural Information Processing Systems 2. Morgan Kaufman: San Francisco.Google Scholar
  30. Mainen ZF, Carnevale NT, Zador AM, Claiborne BJ, Brown TH (1996) Electrotonic architecture of hippocampal CA1 pyramidal neurons based on three-dimensional reconstructions. J. Neurophysiol. 76: 1904-1923.Google Scholar
  31. Mainen ZF, Sejnowski TJ (1996) Influence of dendritic structure on firing pattern in model neocortical neurons. Nature 382: 363-366.Google Scholar
  32. Markram H, Lübke J, Frotscher M, Sakmann B (1997) Regulation of synaptic efficacy by coincidence of postsynaptic APs and EPSPs. Science 275: 213-215.Google Scholar
  33. Markram H, Wang Y, Tsodyks M (1998) Differential signaling via the same axon of neocortical pyramidal neurons. Proc. Natl. Acad. Sci. USA 98: 5323-5328.Google Scholar
  34. Mazzoni P, Andersen RA, Jordan MI (1991) A more biologically plausible learning rule for neural networks. Proc. Natl. Acad. Sci. USA 88: 4433-4437.Google Scholar
  35. Mel BW (1993) Synaptic integration in an excitable dendritic tree. J. Neurophysiol. 70: 1086-1101.Google Scholar
  36. Mel BW, Ruderman DL, Archie KA (1998) Translation-invariant orientation tuning in visual “complex” cells could derive from intradendritic computations. J. Neurosci. 18: 4325-4334.Google Scholar
  37. Minsky M, Papert S (1969) Perceptrons: An Introduction to Computational Geometry. MIT Press: Cambridge, MA.Google Scholar
  38. O'Reilly RC (1996) Biologically plausible error-driven learning using local activation differences: The generalized recirculation algorithm. Neur. Comput. 8: 895-938.Google Scholar
  39. Pike FG, Meredith RM, Olding AWA, Paulsen O (1999) Postsynaptic bursting is essential for “Hebbian” induction of associative long-term potentiation at excitatory synapses in rat hippocampus. J. Phys. 518: 571-576.Google Scholar
  40. Rumelhart D, McCleland J (1986) Parallel Distributed Processing. Bradford Books, Cambridge, MA.Google Scholar
  41. Salin PA, Bullier J (1995) Corticocortical connections in the visual system: Structure and function. Physiol. Rev 75: 107-154.Google Scholar
  42. Schäfer AT, Roth A, Sakmann B (2000) Morphological correlates of BAC-firing threshold in model layer 5 pyramidal neurons. Forum of European Neuroscience Brighton. Eur. J. Neurosci. Supp. 167.08.Google Scholar
  43. Schiller J, Major G, Koester HJ, Schiller Y (2000) NMDA spikes in basal dendrites of cortical pyramidal neurons. Nature 404: 285-289.Google Scholar
  44. Schiller J, Schiller Y, Stuart G, Sakmann B (1997) Calcium action potentials restricted to distal apical dendrites of rat neocortical pyramidal neurons. J. Physiol. (Lond.) 505: 605-616.Google Scholar
  45. Siegel M, Körding KP, König P (2000) Integrating bottom-up and top-down sensory processing by somato-dendritic interactions. J. Comput. Neurosci. 8: 161-173.Google Scholar
  46. Singer W, Gray CM (1995) Visual feature integration and the temporal correlation hypothesis. Ann. Rev. Neurosci. 18: 555-586.Google Scholar
  47. Softky W (1994) Submillisecond coincidence detection in active dendritic trees. Neuroscience 58: 13-41.Google Scholar
  48. Stone JV, Bray AJ (1995). A learning rule for extracting spatiotemporal invariances. Network: Comput. Neural Syst. 6: 429-436.Google Scholar
  49. Stuart GJ, Sakmann B (1994) Active propagation of somatic action potentials into neocortical pyramidal cell dendrites. Nature 367: 69-72.Google Scholar
  50. Stuart GJ, Schiller J, Sakmann B (1997) Action potential initiation and propagation in rat neocortical pyramidal neurons. J. Physiol. (Lond.) 505: 617-632.Google Scholar
  51. Tesauro G (1990) Neural models of classical conditioning: A theoretical viewpoint. In: SJ, Hanson CR Olson, eds. Connectionist Modeling and Brain Function. MIT Press: Cambridge, MA.Google Scholar
  52. Tsodyks MV, Markram H (1997) The neural code between neocortical pyramidal neurons depends on neurotransmitter release probability. Proc. Natl. Acad. Sci. USA 94: 719-723.Google Scholar
  53. Werbos PJ (1974/1994) The Roots of Backpropagation. Wiley, New York (includes Ph.D. thesis of Werbos P from 1974).Google Scholar
  54. Zador AM, Agmon-Snir H, Segev I (1995) The morphoelectrotonic transform: Agraphical approach to dendritic function. J. Neurosci. 15: 1669-1682.Google Scholar
  55. Zeki S, Shipp S (1988) The functional logic of cortical connections. Nature 335: 311-317.Google Scholar
  56. Zipser D, Andersen RA (1988) A backpropagation programmed network that simulates response properties of a subset of posterior parietal neurons. Nature 331: 679-684.Google Scholar
  57. Zipser D, Rumelhart DE (1990) Neurobiological significance of new learning models. In: E Schwartz, ed. Computational Neuroscience MIT Press: Cambridge, MA, pp. 192-200.Google Scholar

Copyright information

© Kluwer Academic Publishers 2001

Authors and Affiliations

  • Konrad P. Körding
    • 1
  • Peter König
    • 1
  1. 1.Institute of NeuroinformaticsETH/UNI ZürichZürichSwitzerland

Personalised recommendations