Recurrent autoassociative networks and holistic computations

Presents an experimental study of holistic computations over distributed representations (DRs) of sequences developed by recurrent autoassociative networks (RANs). Three groups of holistic operators are studied: extracting symbols at fixed positions, extracting symbols at variable positions, and reversing strings. Success with those operators and the very good generalization achieved pave the way for holistic linguistic transformations. Also, it brings a better understanding of the structure of the DRs developed by the RANs.