The functional plausibility of topologically extended models of RBMs as hippocampal models

Masahiko Osawa, Michita Imai

Research output: Contribution to journalConference article

1 Citation (Scopus)

Abstract

The hippocampus has distinctive functional and structural properties. In this research, we utilize Restricted Boltzmann Machines (RBMs) based models inspired by distinctive structures found in hippocampus; neurogenesis in the dentate gyrus (DG) and recurrent connections in CA3. We review two types of topologically extended models of RBMs inspired by hippocampus. In one type of models, units are dynamically added during the training phase, and in the other type, connections are partly recursive. We analyzed these models both as separate models and combined model. The two types of the proposed models implement functions that the hippocampus has but the classical RBMs don't. Furthermore, by combining the two proposed models, memorization of chronologically ordered data and memory reconstruction tasks' performance improved significantly.

Original languageEnglish
Pages (from-to)341-346
Number of pages6
JournalProcedia Computer Science
Volume123
DOIs
Publication statusPublished - 2018 Jan 1
Event8th Annual International Conference on Biologically Inspired Cognitive Architectures, BICA 2017 - Moscow, Russian Federation
Duration: 2017 Aug 12017 Aug 6

Fingerprint

Structural properties
Data storage equipment

Keywords

  • Computational neuroscience
  • Hippocampus
  • Restricted Boltzmann machine

ASJC Scopus subject areas

  • Computer Science(all)

Cite this

The functional plausibility of topologically extended models of RBMs as hippocampal models. / Osawa, Masahiko; Imai, Michita.

In: Procedia Computer Science, Vol. 123, 01.01.2018, p. 341-346.

Research output: Contribution to journalConference article

@article{7280069d13844dcc9b740401dde90a2b,
title = "The functional plausibility of topologically extended models of RBMs as hippocampal models",
abstract = "The hippocampus has distinctive functional and structural properties. In this research, we utilize Restricted Boltzmann Machines (RBMs) based models inspired by distinctive structures found in hippocampus; neurogenesis in the dentate gyrus (DG) and recurrent connections in CA3. We review two types of topologically extended models of RBMs inspired by hippocampus. In one type of models, units are dynamically added during the training phase, and in the other type, connections are partly recursive. We analyzed these models both as separate models and combined model. The two types of the proposed models implement functions that the hippocampus has but the classical RBMs don't. Furthermore, by combining the two proposed models, memorization of chronologically ordered data and memory reconstruction tasks' performance improved significantly.",
keywords = "Computational neuroscience, Hippocampus, Restricted Boltzmann machine",
author = "Masahiko Osawa and Michita Imai",
year = "2018",
month = "1",
day = "1",
doi = "10.1016/j.procs.2018.01.053",
language = "English",
volume = "123",
pages = "341--346",
journal = "Procedia Computer Science",
issn = "1877-0509",
publisher = "Elsevier BV",

}

TY - JOUR

T1 - The functional plausibility of topologically extended models of RBMs as hippocampal models

AU - Osawa, Masahiko

AU - Imai, Michita

PY - 2018/1/1

Y1 - 2018/1/1

N2 - The hippocampus has distinctive functional and structural properties. In this research, we utilize Restricted Boltzmann Machines (RBMs) based models inspired by distinctive structures found in hippocampus; neurogenesis in the dentate gyrus (DG) and recurrent connections in CA3. We review two types of topologically extended models of RBMs inspired by hippocampus. In one type of models, units are dynamically added during the training phase, and in the other type, connections are partly recursive. We analyzed these models both as separate models and combined model. The two types of the proposed models implement functions that the hippocampus has but the classical RBMs don't. Furthermore, by combining the two proposed models, memorization of chronologically ordered data and memory reconstruction tasks' performance improved significantly.

AB - The hippocampus has distinctive functional and structural properties. In this research, we utilize Restricted Boltzmann Machines (RBMs) based models inspired by distinctive structures found in hippocampus; neurogenesis in the dentate gyrus (DG) and recurrent connections in CA3. We review two types of topologically extended models of RBMs inspired by hippocampus. In one type of models, units are dynamically added during the training phase, and in the other type, connections are partly recursive. We analyzed these models both as separate models and combined model. The two types of the proposed models implement functions that the hippocampus has but the classical RBMs don't. Furthermore, by combining the two proposed models, memorization of chronologically ordered data and memory reconstruction tasks' performance improved significantly.

KW - Computational neuroscience

KW - Hippocampus

KW - Restricted Boltzmann machine

UR - http://www.scopus.com/inward/record.url?scp=85045618498&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85045618498&partnerID=8YFLogxK

U2 - 10.1016/j.procs.2018.01.053

DO - 10.1016/j.procs.2018.01.053

M3 - Conference article

AN - SCOPUS:85045618498

VL - 123

SP - 341

EP - 346

JO - Procedia Computer Science

JF - Procedia Computer Science

SN - 1877-0509

ER -