A New Approach towards Vision Suggested by Biologically Realistic Neural Microcircuit Models
We propose an alternative paradigm for processing time-varying visual inputs, in particular for tasks involving temporal and spatial integration, which is inspired by hypotheses about the computational role of cortical microcircuits. Since detailed knowledge about the precise structure of the microcircuit is not needed for that, it can in principle also be implemented with partially unknown or faulty analog hardware. In addition, this approach supports parallel real-time processing of time-varying visual inputs for diverse tasks, since different readouts can be trained to extract concurrently from the same microcircuit completely different information components.
KeywordsCortical microcircuits recurrent connections spiking neurons dynamic synapses dynamical systems movement prediction direction of motion novelty detection parallel computing
Unable to display preview. Download preview PDF.
- [Abbott and Blum, 1996]
- [Auer et al., 2002]Auer, P., Burgsteiner, H., and Maass, W. (2002) Reducing communication for distributed learning in neural networks. In Proc. ICANN’2002, 2002. Springer-Verlag.Google Scholar
- [Buonomano and Merzenich, 1995]Buonomano, D. V., and Merzenich, M. M. (1995) Temporal information transformed into a spatial code by a neural network with realistic properties, Science, vol. 267, Feb. 1995, 1028–1030.Google Scholar
- [Gupta et al., 2000]
- [Haeusler et al., 2002]Haeusler, S., Markram, H., and Maass, W. (2002) Observations on low dimensional readouts from the complex high dimensional dynamics of neural microcircuits, submitted for publication. Online available as # 137 on http://www.igi.tugraz.at/maass/publications.html.
- [Jaeger, 2001]Jaeger, H. (2001) The “echo state” approach to analyzing and training recurrent neural networks, submitted for publication.Google Scholar
- [Levy, 1996]
- [Maass et al., 2001]Maass, W., Natschlaeger, T., and Markram, H. (2001) Real-time computing without stable states: A new framework for neural computation based on perturbations, Neural Computation (in press). Online available as # 130 on http://www.igi.tugraz.at/maass/publications.html.
- [Mallot, 2000]Mallot, H. A. (2000) Computational Vision, MIT-Press (Cambridge, MA).Google Scholar
- [Markram et al., 1998]Markram, H., Wang, Y., and Tsodyks, M. (1998) Differential signaling via the same axon of neocortical pyramidal neurons, Proc. Natl. Acad. Sci., 95, 5323–5328.Google Scholar
- [Rao and Sejnowski, 2000]Rao, R. P. N., and Sejnowski, T. J. (2000) Predictive sequence learning in recurrent neocortical circuits, Advances in Neural Information Processing Systems 12, (NIPS*99), 164–170, S. A. Solla, T. K. Leen, and K. R. Muller (Eds.), MIT Press.Google Scholar
- [Schölkopf and Smola, 2002]Schölkopf, B., and Smola, A. J. (2002) Learning with Kernels, MIT-Press (Cambridge, MA).Google Scholar
- [Stocker and Douglas, 1999]Stocker, A., and Douglas, R. (1999) Computation of smooth optical flow in a feedback connected analog network. Advances in Neural Information Processing Systems 11, (NIPS*98), 706–712.Google Scholar
- [Tsodyks et al., 2000]Tsodyks, M., Uziel, A., and Markram, H. (2000) Synchrony generation in recurrent networks with frequency-dependent synapses, J. Neuroscience, Vol. 20, RC50.Google Scholar
- [Vapnik, 1998]