Associative memory of structured knowledge.
Journal
Scientific reports
ISSN: 2045-2322
Titre abrégé: Sci Rep
Pays: England
ID NLM: 101563288
Informations de publication
Date de publication:
17 12 2022
17 12 2022
Historique:
received:
10
07
2022
accepted:
02
12
2022
entrez:
17
12
2022
pubmed:
18
12
2022
medline:
21
12
2022
Statut:
epublish
Résumé
A long standing challenge in biological and artificial intelligence is to understand how new knowledge can be constructed from known building blocks in a way that is amenable for computation by neuronal circuits. Here we focus on the task of storage and recall of structured knowledge in long-term memory. Specifically, we ask how recurrent neuronal networks can store and retrieve multiple knowledge structures. We model each structure as a set of binary relations between events and attributes (attributes may represent e.g., temporal order, spatial location, role in semantic structure), and map each structure to a distributed neuronal activity pattern using a vector symbolic architecture scheme.We then use associative memory plasticity rules to store the binarized patterns as fixed points in a recurrent network. By a combination of signal-to-noise analysis and numerical simulations, we demonstrate that our model allows for efficient storage of these knowledge structures, such that the memorized structures as well as their individual building blocks (e.g., events and attributes) can be subsequently retrieved from partial retrieving cues. We show that long-term memory of structured knowledge relies on a new principle of computation beyond the memory basins. Finally, we show that our model can be extended to store sequences of memories as single attractors.
Identifiants
pubmed: 36528630
doi: 10.1038/s41598-022-25708-y
pii: 10.1038/s41598-022-25708-y
pmc: PMC9759586
doi:
Types de publication
Journal Article
Research Support, N.I.H., Extramural
Research Support, Non-U.S. Gov't
Langues
eng
Sous-ensembles de citation
IM
Pagination
21808Subventions
Organisme : NINDS NIH HHS
ID : U19 NS104653
Pays : United States
Informations de copyright
© 2022. The Author(s).
Références
Psychol Rev. 2017 Nov;124(6):795-860
pubmed: 29106269
Psychon Bull Rev. 2021 Feb;28(1):40-80
pubmed: 32885404
Psychol Rev. 1995 Jul;102(3):419-457
pubmed: 7624455
Phys Rev A Gen Phys. 1985 Aug;32(2):1007-1018
pubmed: 9896156
Neural Comput. 2018 Jun;30(6):1449-1513
pubmed: 29652585
IEEE Trans Neural Netw. 1995;6(3):623-41
pubmed: 18263348
J Neurosci. 2013 Feb 27;33(9):3844-56
pubmed: 23447596
Annu Rev Physiol. 2002;64:355-405
pubmed: 11826273
Psychol Rev. 2020 Oct;127(5):792-828
pubmed: 32191075
Proc Natl Acad Sci U S A. 1982 Apr;79(8):2554-8
pubmed: 6953413
Cogn Neurodyn. 2013 Feb;7(1):1-11
pubmed: 24427186
Commun Integr Biol. 2011 Sep;4(5):543-8
pubmed: 22046457
Nat Rev Neurosci. 2010 Jul;11(7):474-89
pubmed: 20531421
Phys Rev A Gen Phys. 1987 Jan 1;35(1):380-392
pubmed: 9897963
Psychol Rev. 2018 Jul;125(4):453-485
pubmed: 29952620
Phys Rev Lett. 1985 Sep 30;55(14):1530-1533
pubmed: 10031847
Neuron. 2018 Oct 24;100(2):490-509
pubmed: 30359611
Elife. 2021 Jul 14;10:
pubmed: 34259626
Psychol Rev. 2020 Apr;127(3):327-361
pubmed: 32223284
Cell. 2020 Nov 25;183(5):1249-1263.e23
pubmed: 33181068
Behav Brain Sci. 2003 Dec;26(6):651-65; discussion 666-707
pubmed: 15377127
IEEE Trans Neural Netw Learn Syst. 2021 Sep 03;PP:
pubmed: 34478381
Psychol Rev. 2021 Nov;128(6):1197-1205
pubmed: 34570522
Psychol Rev. 1948 Jul;55(4):189-208
pubmed: 18870876
Neural Comput. 2013 Aug;25(8):1994-2037
pubmed: 23663149
J Neurosci. 2005 Oct 26;25(43):9968-77
pubmed: 16251445
Learn Mem. 2009 Sep 30;16(10):616-24
pubmed: 19794187
Trends Cogn Sci. 2002 Dec 1;6(12):505-510
pubmed: 12475710
Neural Netw. 1999 Jul;12(6):869-876
pubmed: 12662662
Hippocampus. 1994 Dec;4(6):661-82
pubmed: 7704110
Phys Rev Lett. 1986 Dec 1;57(22):2861-2864
pubmed: 10033885