Abstract
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 |
Pages | 601-608 |
Number of pages | 8 |
State | Published - Dec 1 1988 |
Externally published | Yes |
All Science Journal Classification (ASJC) codes
- Engineering(all)