On-chip face recognition system design with memristive Hierarchical Temporal Memory

Timur Ibrayev, Ulan Myrzakhan, Olga Krestinskaya, Aidana Irmanova, Alex Pappachen James

Research output: Contribution to journalArticle

4 Citations (Scopus)

Abstract

Hierarchical Temporal Memory is a new machine learning algorithm intended to mimic the working principle of the neocortex, part of the human brain, responsible for learning, classification, and making predictions. Although many works illustrate its effectiveness as a software algorithm, hardware design for HTM remains an open research problem. Hence, this work proposes an architecture for HTM Spatial Pooler and Temporal Memory with learning mechanism, which creates a single image for each class based on important and unimportant features of all images in the training set. In turn, the reduction in the number of templates within database reduces the memory requirements and increases the processing speed. Moreover, face recognition analysis indicates that for a large number of training images, the proposed design provides higher accuracy results (83.5%) compared to only Spatial Pooler design presented in the previous works.

Original languageEnglish
Pages (from-to)1393-1402
Number of pages10
JournalJournal of Intelligent and Fuzzy Systems
Volume34
Issue number3
DOIs
Publication statusPublished - Jan 1 2018
Externally publishedYes

Fingerprint

Face recognition
Face Recognition
System Design
Chip
Systems analysis
Data storage equipment
Hardware Design
Learning algorithms
Learning systems
Template
Learning Algorithm
Brain
Machine Learning
High Accuracy
Hardware
Software
Prediction
Requirements
Processing
Training

Keywords

  • face recognition
  • HTM
  • memristor
  • spatial pooler
  • temporal memory

ASJC Scopus subject areas

  • Statistics and Probability
  • Engineering(all)
  • Artificial Intelligence

Cite this

On-chip face recognition system design with memristive Hierarchical Temporal Memory. / Ibrayev, Timur; Myrzakhan, Ulan; Krestinskaya, Olga; Irmanova, Aidana; James, Alex Pappachen.

In: Journal of Intelligent and Fuzzy Systems, Vol. 34, No. 3, 01.01.2018, p. 1393-1402.

Research output: Contribution to journalArticle

Ibrayev, Timur ; Myrzakhan, Ulan ; Krestinskaya, Olga ; Irmanova, Aidana ; James, Alex Pappachen. / On-chip face recognition system design with memristive Hierarchical Temporal Memory. In: Journal of Intelligent and Fuzzy Systems. 2018 ; Vol. 34, No. 3. pp. 1393-1402.
@article{a8c69fcadc27452a895fdd33871654fa,
title = "On-chip face recognition system design with memristive Hierarchical Temporal Memory",
abstract = "Hierarchical Temporal Memory is a new machine learning algorithm intended to mimic the working principle of the neocortex, part of the human brain, responsible for learning, classification, and making predictions. Although many works illustrate its effectiveness as a software algorithm, hardware design for HTM remains an open research problem. Hence, this work proposes an architecture for HTM Spatial Pooler and Temporal Memory with learning mechanism, which creates a single image for each class based on important and unimportant features of all images in the training set. In turn, the reduction in the number of templates within database reduces the memory requirements and increases the processing speed. Moreover, face recognition analysis indicates that for a large number of training images, the proposed design provides higher accuracy results (83.5{\%}) compared to only Spatial Pooler design presented in the previous works.",
keywords = "face recognition, HTM, memristor, spatial pooler, temporal memory",
author = "Timur Ibrayev and Ulan Myrzakhan and Olga Krestinskaya and Aidana Irmanova and James, {Alex Pappachen}",
year = "2018",
month = "1",
day = "1",
doi = "10.3233/JIFS-169434",
language = "English",
volume = "34",
pages = "1393--1402",
journal = "Journal of Intelligent and Fuzzy Systems",
issn = "1064-1246",
publisher = "IOS Press",
number = "3",

}

TY - JOUR

T1 - On-chip face recognition system design with memristive Hierarchical Temporal Memory

AU - Ibrayev, Timur

AU - Myrzakhan, Ulan

AU - Krestinskaya, Olga

AU - Irmanova, Aidana

AU - James, Alex Pappachen

PY - 2018/1/1

Y1 - 2018/1/1

N2 - Hierarchical Temporal Memory is a new machine learning algorithm intended to mimic the working principle of the neocortex, part of the human brain, responsible for learning, classification, and making predictions. Although many works illustrate its effectiveness as a software algorithm, hardware design for HTM remains an open research problem. Hence, this work proposes an architecture for HTM Spatial Pooler and Temporal Memory with learning mechanism, which creates a single image for each class based on important and unimportant features of all images in the training set. In turn, the reduction in the number of templates within database reduces the memory requirements and increases the processing speed. Moreover, face recognition analysis indicates that for a large number of training images, the proposed design provides higher accuracy results (83.5%) compared to only Spatial Pooler design presented in the previous works.

AB - Hierarchical Temporal Memory is a new machine learning algorithm intended to mimic the working principle of the neocortex, part of the human brain, responsible for learning, classification, and making predictions. Although many works illustrate its effectiveness as a software algorithm, hardware design for HTM remains an open research problem. Hence, this work proposes an architecture for HTM Spatial Pooler and Temporal Memory with learning mechanism, which creates a single image for each class based on important and unimportant features of all images in the training set. In turn, the reduction in the number of templates within database reduces the memory requirements and increases the processing speed. Moreover, face recognition analysis indicates that for a large number of training images, the proposed design provides higher accuracy results (83.5%) compared to only Spatial Pooler design presented in the previous works.

KW - face recognition

KW - HTM

KW - memristor

KW - spatial pooler

KW - temporal memory

UR - http://www.scopus.com/inward/record.url?scp=85044751478&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85044751478&partnerID=8YFLogxK

U2 - 10.3233/JIFS-169434

DO - 10.3233/JIFS-169434

M3 - Article

VL - 34

SP - 1393

EP - 1402

JO - Journal of Intelligent and Fuzzy Systems

JF - Journal of Intelligent and Fuzzy Systems

SN - 1064-1246

IS - 3

ER -