MicroARTMAP: Use of Mutual Information for Category Reduction in Fuzzy ARTMAP

@inproceedings{GmezSnchez2000MicroARTMAPUO,
  title={MicroARTMAP: Use of Mutual Information for Category Reduction in Fuzzy ARTMAP},
  author={Eduardo G{\'o}mez-S{\'a}nchez and Yannis A. Dimitriadis and J. Manuel Cano Izquierdo and Juan L{\'o}pez Coronado},
  booktitle={IJCNN},
  year={2000}
}
A new architecture, called MicroARTMAP, is proposed to impact the category proliferation problem present in Fuzzy ARTMAP. It handles probabilistic information through the optimization of the mutual information between the input and output spaces, but allowing a small training error, thus avoiding over tting. While reducing the number of categories used by Fuzzy ARTMAP, it holds several desirable properties, such as a correct treatment of exceptions and a fast algorithm, as opposed to other… CONTINUE READING