Guest Talk: Giacomo Indiveri

26. Januar 2015
CITEC Lecture Hall

Title: Event-based mixed-signal electronic circuits for neuromorphic computing systems

Abstract: Neuromorphic computing aims to reproduce the principles of neural computation by emulating as faithfully as possible the detailed biophysics of the nervous system in hardware. A major characteristic of these microelectronic systems is their use of transistors operated inthe ``weak inversion'' or ``subthreshold'' domain. In this domain, the behavior of the transistors is governed by the same Boltzman statistics that characterize the properties of proteic channels in neuron cell membranes. Other important characteristics of neuromorphic systems include the use of ``spikes'' for representing and processing signals, and the presence of learning and adaptation at multiple time scales. In this presentation I will describe neuromorphic electronic circuits that directly emulate the properties of neurons and synapses, and show how they can be configured to implement real-time compact neural processing systems. I will describe hardware models of spiking neurons, synapses, including spike-based learning and plasticity mechanisms, and show how to configure networks of such circuits to implement neuromorphic cognitive systems.