Flexible kernel memory.
This paper introduces a new model of associative memory, capable of both binary and continuous-valued inputs. Based on kernel theory, the memory model is on one hand a generalization of Radial Basis Function networks and, on the other, is in feature space, analogous to a Hopfield network. Attractors...
Guardado en:
Autores principales: | , |
---|---|
Formato: | article |
Lenguaje: | EN |
Publicado: |
Public Library of Science (PLoS)
2010
|
Materias: | |
Acceso en línea: | https://doaj.org/article/1df67ecceedb4f71842f5a9388e702b1 |
Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
id |
oai:doaj.org-article:1df67ecceedb4f71842f5a9388e702b1 |
---|---|
record_format |
dspace |
spelling |
oai:doaj.org-article:1df67ecceedb4f71842f5a9388e702b12021-12-02T20:21:00ZFlexible kernel memory.1932-620310.1371/journal.pone.0010955https://doaj.org/article/1df67ecceedb4f71842f5a9388e702b12010-06-01T00:00:00Zhttps://www.ncbi.nlm.nih.gov/pmc/articles/pmid/20552013/pdf/?tool=EBIhttps://doaj.org/toc/1932-6203This paper introduces a new model of associative memory, capable of both binary and continuous-valued inputs. Based on kernel theory, the memory model is on one hand a generalization of Radial Basis Function networks and, on the other, is in feature space, analogous to a Hopfield network. Attractors can be added, deleted, and updated on-line simply, without harming existing memories, and the number of attractors is independent of input dimension. Input vectors do not have to adhere to a fixed or bounded dimensionality; they can increase and decrease it without relearning previous memories. A memory consolidation process enables the network to generalize concepts and form clusters of input data, which outperforms many unsupervised clustering techniques; this process is demonstrated on handwritten digits from MNIST. Another process, reminiscent of memory reconsolidation is introduced, in which existing memories are refreshed and tuned with new inputs; this process is demonstrated on series of morphed faces.Dimitri NowickiHava SiegelmannPublic Library of Science (PLoS)articleMedicineRScienceQENPLoS ONE, Vol 5, Iss 6, p e10955 (2010) |
institution |
DOAJ |
collection |
DOAJ |
language |
EN |
topic |
Medicine R Science Q |
spellingShingle |
Medicine R Science Q Dimitri Nowicki Hava Siegelmann Flexible kernel memory. |
description |
This paper introduces a new model of associative memory, capable of both binary and continuous-valued inputs. Based on kernel theory, the memory model is on one hand a generalization of Radial Basis Function networks and, on the other, is in feature space, analogous to a Hopfield network. Attractors can be added, deleted, and updated on-line simply, without harming existing memories, and the number of attractors is independent of input dimension. Input vectors do not have to adhere to a fixed or bounded dimensionality; they can increase and decrease it without relearning previous memories. A memory consolidation process enables the network to generalize concepts and form clusters of input data, which outperforms many unsupervised clustering techniques; this process is demonstrated on handwritten digits from MNIST. Another process, reminiscent of memory reconsolidation is introduced, in which existing memories are refreshed and tuned with new inputs; this process is demonstrated on series of morphed faces. |
format |
article |
author |
Dimitri Nowicki Hava Siegelmann |
author_facet |
Dimitri Nowicki Hava Siegelmann |
author_sort |
Dimitri Nowicki |
title |
Flexible kernel memory. |
title_short |
Flexible kernel memory. |
title_full |
Flexible kernel memory. |
title_fullStr |
Flexible kernel memory. |
title_full_unstemmed |
Flexible kernel memory. |
title_sort |
flexible kernel memory. |
publisher |
Public Library of Science (PLoS) |
publishDate |
2010 |
url |
https://doaj.org/article/1df67ecceedb4f71842f5a9388e702b1 |
work_keys_str_mv |
AT dimitrinowicki flexiblekernelmemory AT havasiegelmann flexiblekernelmemory |
_version_ |
1718374152433303552 |