Abstract

Artificial neural networks are notoriously power- and time-consuming when implemented on conventional von Neumann computing systems. Consequently, recent years have seen an emergence of research in machine learning hardware that strives to bring memory and computing closer together. A popular approach is to realise artificial neural networks in hardware by implementing their synaptic weights using memristive devices. However, various device- and system-level non-idealities usually prevent these physical implementations from achieving high inference accuracy. We suggest applying a well-known concept in computer science—committee machines—in the context of memristor-based neural networks. Using simulations and experimental data from three different types of memristive devices, we show that committee machines employing ensemble averaging can successfully increase inference accuracy in physically implemented neural networks that suffer from faulty devices, device-to-device variability, random telegraph noise and line resistance. Importantly, we demonstrate that the accuracy can be improved even without increasing the total number of memristors.

Designing reliable and energy-efficient memristor-based artificial neural networks remains a challenge. Here, the authors demonstrate a technology-agnostic approach, committee machines, which increases the inference accuracy of memristive neural networks that suffer from device variability, faulty devices, random telegraph noise and line resistance.

Details

Title
Committee machines—a universal method to deal with non-idealities in memristor-based neural networks
Author
Joksas, D 1   VIAFID ORCID Logo  ; Freitas, P 2 ; Chai, Z 2 ; Ng, W H 1   VIAFID ORCID Logo  ; Buckwell, M 1 ; Li C 3 ; Zhang, W D 2   VIAFID ORCID Logo  ; Xia, Q 3   VIAFID ORCID Logo  ; Kenyon, A J 1 ; Mehonic, A 1   VIAFID ORCID Logo 

 University College London, Roberts Building, Torrington Place, Department of Electronic and Electrical Engineering, London, UK (GRID:grid.83440.3b) (ISNI:0000000121901201) 
 Liverpool John Moores University, Liverpool, James Parsons Building, Byrom Street, Department of Electronics and Electrical Engineering, Liverpool, UK (GRID:grid.4425.7) (ISNI:0000 0004 0368 0654) 
 University of Massachusetts Amherst, 100 Natural Resources Road, Department of Electrical and Computer Engineering, Amherst, USA (GRID:grid.266683.f) (ISNI:0000 0001 2184 9220) 
Publication year
2020
Publication date
2020
Publisher
Nature Publishing Group
e-ISSN
20411723
Source type
Scholarly Journal
Language of publication
English
ProQuest document ID
2437278191
Copyright
© The Author(s) 2020. This work is published under http://creativecommons.org/licenses/by/4.0/ (the “License”). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.