Work

Hierarchical Temporal Memory Based on Spin-Neurons and Resistive Memory for Energy-Efficient Brain-Inspired Computing

Year: 2015

Type: article

Abstract: Hierarchical temporal memory (HTM) tries to mimic the computing in cerebral neocortex. It identifies spatial and temporal patterns in the input for making inferences. This may require a large number of computationally expensive tasks, such as dot product evaluations. Nanodevices that can provide direct mapping for such primitives are of great interest. In this paper, we propose that the computing blocks for HTM can be mapped using low-voltage, magnetometallic spin-neurons combined with an emerging resistive crossbar network, which involves a comprehensive design at algorithm, architecture, circuit, and device levels. Simulation results show the possibility of more than 200× lower energy as compared with a 45-nm CMOS ASIC design.

Source: IEEE Transactions on Neural Networks and Learning Systems

DOI: 10.1109/tnnls.2015.2462731

Authors: Deliang Fan, Mrigank Sharad, Abhronil Sengupta, Kaushik Roy

Institution: Purdue University West Lafayette

Language: English

Topic: Advanced Memory and Neural Computing

Subfield: Electrical and Electronic Engineering

Field: Engineering

Domain: Physical Sciences

Open Access status: green

Funders: National Science Foundation, Semiconductor Research Corporation, Intel Corporation, Defense Advanced Research Projects Agency

OpenAlex ID: W86197152

FWCI: 5.128

Cites: 53

Cited by: 67

Related to: 10