This page is also offered in these languages:
.

Millions of years of evolution have optimized biological systems with regard to their interactions with the environment they have developed in. In particular simple organisms, such as insects, are remarkably successful and efficient in these interactions, using a minimum of power and computational resources. Recently, humans have started to build artificial perceptive systems that are supposed to autonomously interact with similar environments. The approaches that have commonly been taken to implement such systems have been derived from the same paradigm that has been used to develop the digital computer. The computer has proven itself to be an immensely useful device for the applications it was originally designed for, namely to perform high-speed transformations on unambiguously specified digital data sets. However, artificial perceptive systems that are based on this paradigm are not sharing this success. It has gradually been recognized that the failure of these systems to meet the initial high expectations is partially due to the unsolved problem of efficiently transforming typical sensory input data, which is analog, noisy, and ambiguous, into a format that is suitable for classical computing approaches. This finding has sparked the interest in a newly emerging discipline of bio-inspired engineering, known as Neuromorphic Engineering. This discipline exploits results from the field of NeuroInformatics, whose goal it is to understand the principles of information processing in biological neuronal systems. Neuromorphic Engineering deals with the design and fabrication of artificial neural systems, whose architecture and design principles are based on those of biological nervous systems. Typical examples of components of neuromorphic systems are silicon retinas and cochleas, early visual processing neurons and neural networks, spike-based asynchronous communication/processing infrastructures, learning synapses using analog floating-gate technology, cortical neural network chips, and robotic platforms. Institute of Neuroinformatics University/ETH Zurich Winterthurerstr. 190 CH-8057 Zürich Sweitzerland Phone : +41 1 635 3052 Fax :+41 1 635 3053 http://www.ini.unizh.ch http://www.ini.unizh.ch/satw02