Investigation of Incremental Learning as Temporal Feature Extraction

Shoya Matsumori, Yuki Abe, Masahiko Osawa, Michita Imai

Research output: Contribution to journalConference article

Abstract

In this paper we discuss an effect of feature extraction using Incremental Learning Restricted Boltzmann Machine (IL-RBM). We trained the model on Moving MNIST and analyzed the obtained representation by visualizing hidden activities and reported some meaningful features obtained in incremental learning, similar to that of obtained in Slow Feature Analysis (SFA).

Original languageEnglish
Pages (from-to)342-347
Number of pages6
JournalProcedia Computer Science
Volume145
DOIs
Publication statusPublished - 2018 Jan 1
Event9th Annual International Conference on Biologically Inspired Cognitive Architectures, BICA 2018 - Prague, Czech Republic
Duration: 2018 Aug 222018 Aug 24

Fingerprint

Feature extraction

Keywords

  • incremental learning restricted boltzmann machine
  • slow feature analysis
  • time series analysis

ASJC Scopus subject areas

  • Computer Science(all)

Cite this

Investigation of Incremental Learning as Temporal Feature Extraction. / Matsumori, Shoya; Abe, Yuki; Osawa, Masahiko; Imai, Michita.

In: Procedia Computer Science, Vol. 145, 01.01.2018, p. 342-347.

Research output: Contribution to journalConference article

Matsumori, Shoya ; Abe, Yuki ; Osawa, Masahiko ; Imai, Michita. / Investigation of Incremental Learning as Temporal Feature Extraction. In: Procedia Computer Science. 2018 ; Vol. 145. pp. 342-347.
@article{faacaac179884acebbde91bbf4c0a163,
title = "Investigation of Incremental Learning as Temporal Feature Extraction",
abstract = "In this paper we discuss an effect of feature extraction using Incremental Learning Restricted Boltzmann Machine (IL-RBM). We trained the model on Moving MNIST and analyzed the obtained representation by visualizing hidden activities and reported some meaningful features obtained in incremental learning, similar to that of obtained in Slow Feature Analysis (SFA).",
keywords = "incremental learning restricted boltzmann machine, slow feature analysis, time series analysis",
author = "Shoya Matsumori and Yuki Abe and Masahiko Osawa and Michita Imai",
year = "2018",
month = "1",
day = "1",
doi = "10.1016/j.procs.2018.11.082",
language = "English",
volume = "145",
pages = "342--347",
journal = "Procedia Computer Science",
issn = "1877-0509",
publisher = "Elsevier BV",

}

TY - JOUR

T1 - Investigation of Incremental Learning as Temporal Feature Extraction

AU - Matsumori, Shoya

AU - Abe, Yuki

AU - Osawa, Masahiko

AU - Imai, Michita

PY - 2018/1/1

Y1 - 2018/1/1

N2 - In this paper we discuss an effect of feature extraction using Incremental Learning Restricted Boltzmann Machine (IL-RBM). We trained the model on Moving MNIST and analyzed the obtained representation by visualizing hidden activities and reported some meaningful features obtained in incremental learning, similar to that of obtained in Slow Feature Analysis (SFA).

AB - In this paper we discuss an effect of feature extraction using Incremental Learning Restricted Boltzmann Machine (IL-RBM). We trained the model on Moving MNIST and analyzed the obtained representation by visualizing hidden activities and reported some meaningful features obtained in incremental learning, similar to that of obtained in Slow Feature Analysis (SFA).

KW - incremental learning restricted boltzmann machine

KW - slow feature analysis

KW - time series analysis

UR - http://www.scopus.com/inward/record.url?scp=85059476753&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85059476753&partnerID=8YFLogxK

U2 - 10.1016/j.procs.2018.11.082

DO - 10.1016/j.procs.2018.11.082

M3 - Conference article

AN - SCOPUS:85059476753

VL - 145

SP - 342

EP - 347

JO - Procedia Computer Science

JF - Procedia Computer Science

SN - 1877-0509

ER -