Learning in a neural network model in real time using real world stimuli

In this paper we present a model of the auditory system that is trained usi ng real-world stimuli and running in real-time. The system consists of diff erent sound sources, a microphone, an AID board, a peripheral auditory syst em implemented in software and a central network of spiking neurons. The sy napses formed by peripheral neurons on the central ones are subject to syna ptic plasticity. We implemented a learning rule that depends on the precise temporal relation of pre- and post-synaptic action potentials. We demonstr ate that this mechanism allows the development of receptive fields combinin g learning in real-time, learning with few stimulus presentations and robus t learning in the presence of large imbalances in the probability of occurr ence of individual stimuli. (C) 2001 Elsevier Science B.V. All rights reser ved.