Hierarchical systems use schemata (knowledge sources) to represent knowledge of the environment but it is difficult for them to deal with the variability of the observed data. The authors describe a hierarchical system that uses the hidden Markov model (HMM) methodology to represent both general knowledge about objects and knowledge about their possible instantiations. The HMM is shown to be a compact, computationally efficient and accurate knowledge source. The authors discuss the algorithms used and their implementation using systolic arrays.
|Original language||English (US)|
|Title of host publication||IEEE Int Conf on Neural Networks|
|Publisher||Publ by IEEE|
|Number of pages||8|
|State||Published - Dec 1 1988|
All Science Journal Classification (ASJC) codes