Navigation auf uzh.ch
Navigation auf uzh.ch
The cerebral cortex is a remarkable computational system. It uses slow, unreliable and inhomogeneous computing elements, and yet it outperforms the most powerful computers in relatively routine functions such as, for example, vision. The disparity between the effectiveness of computation in cortical circuits systems and in a computer is primarily attributable to the way the elementary devices are used in the system, and to the kind of computational primitives they implement. We study spike based plasticity, cortical inhibition, recurrent excitation, attractor networks, and soft Winner-Take-All (WTA) networks as mechanisms that underlie these computational primitives. In particular, we carry out simulation studies of the computational properties of networks of spiking neurons that implement such computational primitives using the Brian spiking neural network simulator, and we emulate such networks in real-time using VLSI devices that contain both subthreshold analog, and asynchronous digital, microelectronic circuits. These circuits implement neural and synaptic dynamics with biophysically realistic time constants, and plasticity mechanisms, including Short Term Depression and Spike-Timing Dependent Plasticity (STDP) forms of learning.
Both software simulations and hardware emulations are used to aid our understanding of basic principles of neural computation such as working memory, decision making, pattern recognition, and sequence learning.
