Special issue on fundamental and information processing aspects of neurocomputing

Since its first occurence in 1993, the European Symposium on Artificial Neural Networks ESANN has become a reference for research on fundamental and theoretical aspects of artificial neural networks. Each year, about 100 specialists from all parts of the world attend and contribute to ESANN. The eighth ESANN symposium was organized in the medieval city of Bruges in Belgium from 26 to 28 April 2001. In total, 67 contributed or invited papers were presented, that were all reviewed. Within the broad array of research, there was a particularly strong collection of papers related to the applications of mathematics and statistics to the learning, modelling, data analysis and information processing. Based on these reviews and on proposals by the scientific committee, the authors of many high quality papers were invited to prepare a journal version of their contribution. These papers were thoroughly reviewed by three independent experts. Finally, 22 papers were accepted for this special issue of Neurocomputing on fundamental and information processing aspects of artificial neural networks. In this special issue, the 22 papers are grouped around six main topics: algorithms and learning, model selection, data analysis, forecasting, vision and signal, and image processing