Discrete-level memristive circuits for HTM-based spatiotemporal data classification system
Loading...
Date
Journal Title
Journal ISSN
Volume Title
Publisher
Institution of Engineering and Technology (IET)
Abstract
The authors propose a discrete-level memristive memory design for analogue data processing in hardware implementations of hierarchical temporal memory (HTM). Memristors operate in ternary and quaternary states via sub-cell resistive networks. Integration into HTM temporal memory improves face and speech recognition accuracy.
Description
Citation
Irmanova, Aidana; Ibrayev, Timur; James, Alex Pappachen (2017). Discrete-level memristive circuits for HTM-based spatiotemporal data classification system. IET Cyber-Physical Systems: Theory & Applications, 3(1), 34–43. DOI: 10.1049/iet-cps.2017.0053