An event-based VLSI network of integrate-and-fire neurons

The growing interest in pulse-based neural networks is encouraging the development of hardware implementations of massively parallel, distributed networks of integrate-and-fire (I&F) neurons. We have developed a mixed-mode (analog/digital) VLSI device that comprises a reconfigurable network of I&F neurons and adaptive synapses. The synapses receive input spikes and the neurons transmit output spikes (events) using an asynchronous address-event representation (AER). We describe the network architecture, present experimental data demonstrating the characteristics of the single elements on the chip, and show that a competitive network configuration has winner-take-all (WTA) behaviour and produces spike synchronization.