investigate the ability of neural networks in approximating input-output random functions and transformations, thus widening the range of applicability of such networks;
develop a statistical representation of synthetic and biological signals by exploiting the properties of different universal approximators of stochastic processes, and apply this result to the problems of signal recognition and synthesis;
investigate and develop techniques suitable for the supervised or unsupervised statistical identification of non-stationary non-linear systems.
Laboratory: Artificial Intelligence and Multimedia Applications Lab. at DII
