article · 1992
Sparse Adaptive Memory
M. J. Flynn, Brian Flachs, Brian Flachs, Michael J. Flynn
Cited by 0Abstract and details only
Abstract
Pattern recognition is a budding field with many possible approaches. This paper proposes "Sparse Adaptive Memory," a classifier architecture built upon the strengths of Parzen classifiers, nearest neighbor classifiers, and feed forward neural networks. The structure of Sparse Adaptive Memory is able emulate these established techniques. A key feature of this learning architecture is the ability to adaptively change its prototype patterns. Many interesting properties of this architecture are examined and shown to be related to the underlying probability density surfaces. SAM is described in terms of convolution, sampling and convolution operations applied to these surfaces. As SAM changes the prototype patterns in the list, it isolates modes in the density functions to produce a classifier that is in some senses optimal. Key Words: pattern recognition, adaptive learning, bayesian classification. Table of Contents 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . ....
Cite this paper
Flynn, M. J., Flachs, B., Flachs, B., & Flynn, M. J. (1992). Sparse Adaptive Memory.
Flynn, M. J., et al. “Sparse Adaptive Memory.” 1992.
Flynn, M. J., Brian Flachs, Brian Flachs, and Michael J. Flynn. 1992. “Sparse Adaptive Memory.”
Flynn, M.J. et al. (1992) ‘Sparse Adaptive Memory’.
@article{flynn1992sparse,
title = {{Sparse Adaptive Memory}},
author = {Flynn, M. J. and Flachs, Brian and Flachs, Brian and Flynn, Michael J.},
year = {1992}
}Made from the index record. Check names, capitalisation and page numbers against the paper. Need another style? AI for Research has over 10,000.