An Episodic Memory Model with Slow Features Extraction for Topological Map Building

Wei Hong Chin, Yuichiro Toda, Naoyuki Kubota, Jinseok Woo, Chu Kiong Loo

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

This paper presents a new integration model for on-line topological map building with environment slow features detection. The proposed model is formed by the integration of Bayesian Adaptive Resonance Associative Memory (BARAM) and Incremental Slow Feature Analysis (IncSFA). IncSFA incrementally extracts slowly varying features from a rapidly changing input signal. These slow features will be fed to BARAM for environment learning to build a topological map. The explored environment is represented as a set of neurons (nodes) and edges that connecting all nodes. Each neuron represents a distinct place and edges store robot traverse information that leads the robot to travel from one node to another. The proposed model is an unsupervised learning technique that does not require any prior knowledge of what an environment is supposed to be for ease of implementation. The effectiveness of our proposed method is validated by several standardized benchmark datasets.

Original languageEnglish
Title of host publicationMHS 2018 - 2018 29th International Symposium on Micro-NanoMechatronics and Human Science
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781538667927
DOIs
Publication statusPublished - Dec 2018
Event29th International Symposium on Micro-NanoMechatronics and Human Science, MHS 2018 - Nagoya, Japan
Duration: Dec 10 2018Dec 12 2018

Publication series

NameMHS 2018 - 2018 29th International Symposium on Micro-NanoMechatronics and Human Science

Conference

Conference29th International Symposium on Micro-NanoMechatronics and Human Science, MHS 2018
CountryJapan
CityNagoya
Period12/10/1812/12/18

Fingerprint

Memory Model
pattern recognition
Feature Extraction
Feature extraction
Associative Memory
Data storage equipment
associative memory
Neurons
Neuron
Vertex of a graph
Robot
neurons
robots
Robots
learning
Feature Detection
Unsupervised learning
Unsupervised Learning
Learning Environment
Prior Knowledge

ASJC Scopus subject areas

  • Artificial Intelligence
  • Electrical and Electronic Engineering
  • Mechanical Engineering
  • Control and Optimization
  • Instrumentation
  • Human-Computer Interaction

Cite this

Chin, W. H., Toda, Y., Kubota, N., Woo, J., & Loo, C. K. (2018). An Episodic Memory Model with Slow Features Extraction for Topological Map Building. In MHS 2018 - 2018 29th International Symposium on Micro-NanoMechatronics and Human Science [8887062] (MHS 2018 - 2018 29th International Symposium on Micro-NanoMechatronics and Human Science). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/MHS.2018.8887062

An Episodic Memory Model with Slow Features Extraction for Topological Map Building. / Chin, Wei Hong; Toda, Yuichiro; Kubota, Naoyuki; Woo, Jinseok; Loo, Chu Kiong.

MHS 2018 - 2018 29th International Symposium on Micro-NanoMechatronics and Human Science. Institute of Electrical and Electronics Engineers Inc., 2018. 8887062 (MHS 2018 - 2018 29th International Symposium on Micro-NanoMechatronics and Human Science).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Chin, WH, Toda, Y, Kubota, N, Woo, J & Loo, CK 2018, An Episodic Memory Model with Slow Features Extraction for Topological Map Building. in MHS 2018 - 2018 29th International Symposium on Micro-NanoMechatronics and Human Science., 8887062, MHS 2018 - 2018 29th International Symposium on Micro-NanoMechatronics and Human Science, Institute of Electrical and Electronics Engineers Inc., 29th International Symposium on Micro-NanoMechatronics and Human Science, MHS 2018, Nagoya, Japan, 12/10/18. https://doi.org/10.1109/MHS.2018.8887062
Chin WH, Toda Y, Kubota N, Woo J, Loo CK. An Episodic Memory Model with Slow Features Extraction for Topological Map Building. In MHS 2018 - 2018 29th International Symposium on Micro-NanoMechatronics and Human Science. Institute of Electrical and Electronics Engineers Inc. 2018. 8887062. (MHS 2018 - 2018 29th International Symposium on Micro-NanoMechatronics and Human Science). https://doi.org/10.1109/MHS.2018.8887062
Chin, Wei Hong ; Toda, Yuichiro ; Kubota, Naoyuki ; Woo, Jinseok ; Loo, Chu Kiong. / An Episodic Memory Model with Slow Features Extraction for Topological Map Building. MHS 2018 - 2018 29th International Symposium on Micro-NanoMechatronics and Human Science. Institute of Electrical and Electronics Engineers Inc., 2018. (MHS 2018 - 2018 29th International Symposium on Micro-NanoMechatronics and Human Science).
@inproceedings{1c0d01f7387740a6bf9117cc2d53fc24,
title = "An Episodic Memory Model with Slow Features Extraction for Topological Map Building",
abstract = "This paper presents a new integration model for on-line topological map building with environment slow features detection. The proposed model is formed by the integration of Bayesian Adaptive Resonance Associative Memory (BARAM) and Incremental Slow Feature Analysis (IncSFA). IncSFA incrementally extracts slowly varying features from a rapidly changing input signal. These slow features will be fed to BARAM for environment learning to build a topological map. The explored environment is represented as a set of neurons (nodes) and edges that connecting all nodes. Each neuron represents a distinct place and edges store robot traverse information that leads the robot to travel from one node to another. The proposed model is an unsupervised learning technique that does not require any prior knowledge of what an environment is supposed to be for ease of implementation. The effectiveness of our proposed method is validated by several standardized benchmark datasets.",
author = "Chin, {Wei Hong} and Yuichiro Toda and Naoyuki Kubota and Jinseok Woo and Loo, {Chu Kiong}",
year = "2018",
month = "12",
doi = "10.1109/MHS.2018.8887062",
language = "English",
series = "MHS 2018 - 2018 29th International Symposium on Micro-NanoMechatronics and Human Science",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
booktitle = "MHS 2018 - 2018 29th International Symposium on Micro-NanoMechatronics and Human Science",

}

TY - GEN

T1 - An Episodic Memory Model with Slow Features Extraction for Topological Map Building

AU - Chin, Wei Hong

AU - Toda, Yuichiro

AU - Kubota, Naoyuki

AU - Woo, Jinseok

AU - Loo, Chu Kiong

PY - 2018/12

Y1 - 2018/12

N2 - This paper presents a new integration model for on-line topological map building with environment slow features detection. The proposed model is formed by the integration of Bayesian Adaptive Resonance Associative Memory (BARAM) and Incremental Slow Feature Analysis (IncSFA). IncSFA incrementally extracts slowly varying features from a rapidly changing input signal. These slow features will be fed to BARAM for environment learning to build a topological map. The explored environment is represented as a set of neurons (nodes) and edges that connecting all nodes. Each neuron represents a distinct place and edges store robot traverse information that leads the robot to travel from one node to another. The proposed model is an unsupervised learning technique that does not require any prior knowledge of what an environment is supposed to be for ease of implementation. The effectiveness of our proposed method is validated by several standardized benchmark datasets.

AB - This paper presents a new integration model for on-line topological map building with environment slow features detection. The proposed model is formed by the integration of Bayesian Adaptive Resonance Associative Memory (BARAM) and Incremental Slow Feature Analysis (IncSFA). IncSFA incrementally extracts slowly varying features from a rapidly changing input signal. These slow features will be fed to BARAM for environment learning to build a topological map. The explored environment is represented as a set of neurons (nodes) and edges that connecting all nodes. Each neuron represents a distinct place and edges store robot traverse information that leads the robot to travel from one node to another. The proposed model is an unsupervised learning technique that does not require any prior knowledge of what an environment is supposed to be for ease of implementation. The effectiveness of our proposed method is validated by several standardized benchmark datasets.

UR - http://www.scopus.com/inward/record.url?scp=85075023416&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85075023416&partnerID=8YFLogxK

U2 - 10.1109/MHS.2018.8887062

DO - 10.1109/MHS.2018.8887062

M3 - Conference contribution

AN - SCOPUS:85075023416

T3 - MHS 2018 - 2018 29th International Symposium on Micro-NanoMechatronics and Human Science

BT - MHS 2018 - 2018 29th International Symposium on Micro-NanoMechatronics and Human Science

PB - Institute of Electrical and Electronics Engineers Inc.

ER -