An Internal Mechanism for Detecting Parasite Attractors in a Hopfield Network
Abstract
This paper presents a built-in mechanism for automatic detection of prototypes (as opposed to parasite attractors) in a Hopfield network. It has a good statistical performance and avoids host computer overhead for this purpose. This mechanism is based on an internal coding of the prototypes during learning, using cyclic redundancy codes, and leads to an efficient implementation in VLSI. The immediate separation of prototypes from parasite attractors can be used to enhance the autonomy of the networks, aiming at hierarchical multinetwork structures. As an example, the use of such an architecture for the classification of handwritten digits is described.
Cite
Text
Gascuel et al. "An Internal Mechanism for Detecting Parasite Attractors in a Hopfield Network." Neural Computation, 1994. doi:10.1162/NECO.1994.6.5.902Markdown
[Gascuel et al. "An Internal Mechanism for Detecting Parasite Attractors in a Hopfield Network." Neural Computation, 1994.](https://mlanthology.org/neco/1994/gascuel1994neco-internal/) doi:10.1162/NECO.1994.6.5.902BibTeX
@article{gascuel1994neco-internal,
title = {{An Internal Mechanism for Detecting Parasite Attractors in a Hopfield Network}},
author = {Gascuel, Jean-Dominique and Moobed, Bahram and Weinfeld, Michel},
journal = {Neural Computation},
year = {1994},
pages = {902-915},
doi = {10.1162/NECO.1994.6.5.902},
volume = {6},
url = {https://mlanthology.org/neco/1994/gascuel1994neco-internal/}
}