Noisy spiking neurons with temporal coding have more computational power than sigmoidal neurons

W. Maass

Abstract:

We exhibit a novel way of simulating sigmoidal neural nets by networks of noisy spiking neurons in temporal coding. Furthermore it is shown that networks of noisy spiking neurons with temporal coding have a strictly larger computational power than sigmoidal neural nets with the same number of units.

Reference: W. Maass. Noisy spiking neurons with temporal coding have more computational power than sigmoidal neurons. In M. Mozer, M. I. Jordan, and T. Petsche, editors, Advances in Neural Information Processing Systems, volume 9, pages 211-217. MIT Press (Cambridge), 1997.