Ex-situ training of dense memristor crossbar for neuromorphic applications

Raqibul Hasan; Chris Yakopcic; Tarek M. Taha

doi:10.1109/NANOARCH.2015.7180590

2015 IEEE/ACM International Symposium on Nanoscale Architectures (NANOARCH)

Ex-situ training of dense memristor crossbar for neuromorphic applications

Year: 2015, Pages: 75-81

DOI Bookmark: 10.1109/NANOARCH.2015.7180590

Authors

Raqibul Hasan, Department of Electrical and Computer Engineering, University of Dayton, Ohio, USA
Chris Yakopcic, Department of Electrical and Computer Engineering, University of Dayton, Ohio, USA
Tarek M. Taha, Department of Electrical and Computer Engineering, University of Dayton, Ohio, USA

Abstract

This study proposes a technique for programming a dense memristor crossbar array without isolation transistors (0T1M) in order to achieve ex-situ training of a neural network. Programming memristors to a specific resistance level requires an iterative process needing the reading of individual memristor resistances due to memristor device stochasticity. This paper presents a circuit to read individual resistances from a 0T1M crossbar and a method to map neuron synaptic weights into a novel neural circuit to enable ex-situ training. The results show that we are able to train the resistances in a 0T1M crossbar and that the 0T1M system is about 93% smaller in area than 1T1M systems.

Like what you’re reading?

Already a member?

Get this article FREE with a new membership!

A memristor network with coupled oscillator and crossbar towards L2-norm based machine learning
2016 IEEE/ACM International Symposium on Nanoscale Architectures (NANOARCH)
Memristor Crossbar-Based Pattern Recognition Circuit Using Perceptron Learning Rule
2016 IEEE International Symposium on Nanoelectronic and Information Systems (iNIS)
A Memristor Crossbar Based Computing Engine Optimized for High Speed and Accuracy
2016 IEEE Computer Society Annual Symposium on VLSI (ISVLSI)
Interconnect networks for memristor crossbar
2015 IEEE/ACM International Symposium on Nanoscale Architectures (NANOARCH)
Fast boolean logic mapped on memristor crossbar
2015 33rd IEEE International Conference on Computer Design (ICCD)
Robust neural logic block (NLB) based on memristor crossbar array
2011 IEEE/ACM International Symposium on Nanoscale Architectures
Memristor Crossbar Based Programmable Interconnects
2014 IEEE Computer Society Annual Symposium on VLSI (ISVLSI)
A Hardware Architecture for Columnar-Organized Memory Based on CMOS Neuron and Memristor Crossbar Arrays
IEEE Transactions on Very Large Scale Integration (VLSI) Systems
Parasitic-Aware Modelling for Neural Networks Implemented with Memristor Crossbar Array
2021 IEEE 14th International Symposium on Embedded Multicore/Many-core Systems-on-Chip (MCSoC)
CSrram: Area-Efficient Low-Power Ex-Situ Training Framework for Memristive Neuromorphic Circuits Based on Clustered Sparsity
2019 IEEE Computer Society Annual Symposium on VLSI (ISVLSI)

Ex-situ training of dense memristor crossbar for neuromorphic applications

Authors

Abstract

Related Articles