Skip to main content
Log in

Parallel Fiber Coding in the Cerebellum for Life-Long Learning

  • Published:
Autonomous Robots Aims and scope Submit manuscript

Abstract

Continuous and real-time learning is a difficult problem in robotics. To learn efficiently, it is important to recognize the current situation and learn appropriately for that context. To be effective, this requires the integration of a large number of sensorimotor and cognitive signals. So far, few principles on how to perform this integration have been proposed. Another limitation is the difficulty to include the complete contextual information to avoid destructive interference while learning different tasks.

We suggest that a vertebrate brain structure important for sensorimotor coordination, the cerebellum, may provide answers to these difficult problems. We investigate how learning in the input layer of the cerebellum may successfully encode contextual knowledge in a representation useful for coordination and life-long learning. We propose that a sparsely-distributed and statistically-independent representation provides a valid criterion for the self-organizing classification and integration of context signals. A biologically motivated unsupervised learning algorithm that approximate such a representation is derived from maximum likelihood. This representation is beneficial for learning in the cerebellum by simplifying the credit assignment problem between what must be learned and the relevant signals in the current context for learning it. Due to its statistical independence, this representation is also beneficial for life-long learning by reducing the destructive interference across tasks, while retaining the ability to generalize. The benefits of the learning algorithm are investigated in a spiking model that learns to generate predictive smooth pursuit eye movements to follow target trajectories.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Similar content being viewed by others

References

  • Atkeson, C.G., Moore, A.W., and Schaal, S. 1997. Locally weighted learning for control. Artificial Intelligence Review, 11(1–5):75–113.

    Google Scholar 

  • Barbour, B. 1993. Synaptic currents evoked in Purkinje cells by stimulating individual granule cells. Neuron, 11(4):759–769.

    Google Scholar 

  • Blakemore, S.J., Goodbody, S.J., and Wolpert, D.M. 1998. Predicting the consequences of our own actions: The role of sensorimotor context estimation. J Neurosci, 18(18):7511–7518.

    Google Scholar 

  • Blakemore, S.J., Wolpert, D.M., and Frith, C.D. 1999. The cerebellum contributes to somatosensory cortical activity during self-produced tactile stimulation. Neuroimage, 10(4):448–459.

    Google Scholar 

  • Coenen, O.J.-M.D. 1998. Modeling the vestibulo-ocular reflex and the cerebellum: Analytical & computational approaches. Ph.D. thesis, University of California, San Diego. Physics Department. Available at www.cnl.salk.edu/~olivier.

  • Coenen, O.J.-M.D., Arnold, M., Jabri, M.A., Courchesne, E., and Sejnowski, T.J. 1999. A hypothesis for parallel fiber coding in the cerebellum. In Society for Neuroscience Abstracts, Vol. 25.

  • Coenen, O.J.-M.D., Arnold, M.P., Sejnowski, T.J., and Jabri, M.A. 2000. Bayesian analysis for parallel fiber coding in the cerebellum. In Proceedings of the 7th International Conference on Neural Information Processing (ICONIP-2000),Taejon, Korea, pp. 1301–1306.

  • Coenen, O.J.-M.D., Eagleman, D.M., Mitsner, V., Bartol, T.M., Bell, A.J., and Sejnowski, T.J. 2001. Cerebellar glomeruli: Does limited extracellular calcium direct a new kind of plasticity?. In Society for Neuroscience Abstracts, Vol. 27.

  • D'Angelo, E., De Filippi, G., Rossi, P., and Taglietti, V. 1995. Synaptic excitation of individual rat cerebellar granule cells in situ: Evidence for the role of NMDA receptors. J Physiol (Lond), 484 (Pt 2):397–413.

    Google Scholar 

  • D'Angelo, E., Rossi, P., Armano, S., and Taglietti, V. 1999. Evidence for NMDA and mGlu receptor-dependent long-term potentiation of mossy fiber-granule cell transmission in rat cerebellum. J Neurophysiol, 81(1):277–287.

    Google Scholar 

  • Girolami, M., Cichocki, A., and Amari, S.-I. 1998. A Common neural network model for unsupervised exploratory data analysis and independent component analysis. IEEE Transactions on Neural Networks, 9(6):1495.

    Google Scholar 

  • Jakab, R.L. and Hamori, J. 1988. Quantitative morphology and synaptology of cerebellar glomeruli in the rat. Anat Embryol (Berl), 179(1):81–88.

    Google Scholar 

  • Jonker, H.J., Coolen, A.C., and Denier van der Gon, J.J. 1998. Autonomous development of decorrelation filters in neural networks with recurrent inhibition. Network, 9(3):345–362.

    Google Scholar 

  • Kettner, R.E., Mahamud, S., Leung, H.C., Sitkoff, N., Houk, J.C., Peterson, B.W., and G, B.A. 1997. Prediction of complex two-dimensional trajectories by a cerebellar model of smooth pursuit eye movement. Journal of Neurophysiology, 77(4):2115–2130.

    Google Scholar 

  • MacKay, D.J.C. 1996.Maximumlikelihood and covariant algorithms for independent component analysis. Unpublished manuscript, available at http://wol.ra.phy.cam.ac.uk/mackay/BayesICA.html.

  • Marr, D. 1969. A theory of cerebellar cortex. J. Physiol., 202:437–470.

    Google Scholar 

  • Meunier, C. and Nadal, J.-P. 1995. The Handbook of Brain Theory and Neural Networks. Cambridge, MA: MIT Press. pp. 899–901.

    Google Scholar 

  • Miall, R.C., Weir, D.J., Wolpert, D.M., and Stein, J.F. 1993. Is the cerebellum a Smith predictor? Journal of Motor Behavior, 25(3):203–216.

    Google Scholar 

  • Olshausen, B.A. and Field, D.J. 1996. Emergence of simple-cell receptive field properties by learning a sparse code for natural images [see comments]. Nature, 381(6583):607–609.

    Google Scholar 

  • Palay, S.L. and Chan-Palay, V. 1974. Cerebellar Cortex, Cytology and Organization, Springer-Verlag: Berlin.

    Google Scholar 

Download references

Author information

Authors and Affiliations

Authors

Rights and permissions

Reprints and permissions

About this article

Cite this article

Coenen, O.JM., Arnold, M.P., Sejnowski, T.J. et al. Parallel Fiber Coding in the Cerebellum for Life-Long Learning. Autonomous Robots 11, 291–297 (2001). https://doi.org/10.1023/A:1012403510221

Download citation

  • Issue Date:

  • DOI: https://doi.org/10.1023/A:1012403510221

Navigation