Image Based HTM Word Recognizer for Language Processing

Aidana Irmanova, Olga Krestinskaya, Alex James Pappachen

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

Abstract

The hardware implementation of neuro-inspired machine learning algorithms for near sensor processing on edge devices is an open problem. In this work, we propose a solution to written word recognition problem related to sequence learning tasks with images. Applying a theoretical framework of neocortex functionality as a sequence learning algorithm on a hardware implementation of Hierarchical Temporal Memory (HTM), we test the potential use of HTM in near-sensor on-chip natural language processing for text/symbol recognition.

Original languageEnglish
Title of host publication2018 IEEE International Conference on Consumer Electronics - Asia, ICCE-Asia 2018
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781538658079
DOIs
Publication statusPublished - Nov 28 2018
Event2018 IEEE International Conference on Consumer Electronics - Asia, ICCE-Asia 2018 - JeJu, Korea, Republic of
Duration: Jun 24 2018Jun 26 2018

Other

Other2018 IEEE International Conference on Consumer Electronics - Asia, ICCE-Asia 2018
CountryKorea, Republic of
CityJeJu
Period6/24/186/26/18

Keywords

  • HTM
  • Natural language processing
  • Symbol recognition
  • Text recognition

ASJC Scopus subject areas

  • Computer Networks and Communications
  • Electrical and Electronic Engineering
  • Computer Vision and Pattern Recognition

Fingerprint Dive into the research topics of 'Image Based HTM Word Recognizer for Language Processing'. Together they form a unique fingerprint.

Cite this