Hardware implementation of memristor-based artificial neural networks

Fernando Aguirre, Abu Sebastian, Manuel Gallo, Wenhao Song, Tong Wang, J. Joshua Yang, Wei Lu, Meng-Fan Chang, Daniele Ielmini, Yuchao Yang, Adnan Mehonic, Anthony Kenyon, Marco A. Villena, Juan B. Roldán, Yuting Wu, Hung-Hsi Hsu, Nagarajan Raghavan, Jordi Suñé, Enrique Miranda, Ahmed Eltawil, Gianluca Setti, Kamilya Smagulova, Khaled N. Salama, Olga Krestinskaya, Xiaobing Yan, Kah-Wee Ang, Samarth Jain, Sifan Li, Osamah Alharbi, Sebastian Pazos and Mario Lanza ()
Additional contact information
Fernando Aguirre: King Abdullah University of Science and Technology (KAUST)
Abu Sebastian: IBM Research – Zurich
Manuel Gallo: IBM Research – Zurich
Wenhao Song: University of Southern California (USC)
Tong Wang: University of Southern California (USC)
J. Joshua Yang: University of Southern California (USC)
Wei Lu: University of Michigan
Meng-Fan Chang: National Tsing Hua University
Daniele Ielmini: Politecnico di Milano and IUNET
Yuchao Yang: Peking University
Adnan Mehonic: University College London (UCL), Torrington Place
Anthony Kenyon: University College London (UCL), Torrington Place
Marco A. Villena: King Abdullah University of Science and Technology (KAUST)
Juan B. Roldán: Facultad de Ciencias, Universidad de Granada, Avenida Fuentenueva s/n
Yuting Wu: University of Michigan
Hung-Hsi Hsu: National Tsing Hua University
Nagarajan Raghavan: Singapore University of Technology & Design
Jordi Suñé: Universitat Autònoma de Barcelona (UAB)
Enrique Miranda: Universitat Autònoma de Barcelona (UAB)
Ahmed Eltawil: King Abdullah University of Science and Technology (KAUST)
Gianluca Setti: King Abdullah University of Science and Technology (KAUST)
Kamilya Smagulova: King Abdullah University of Science and Technology (KAUST)
Khaled N. Salama: King Abdullah University of Science and Technology (KAUST)
Olga Krestinskaya: King Abdullah University of Science and Technology (KAUST)
Xiaobing Yan: Hebei University
Kah-Wee Ang: National University of Singapore (NUS)
Samarth Jain: National University of Singapore (NUS)
Sifan Li: National University of Singapore (NUS)
Osamah Alharbi: King Abdullah University of Science and Technology (KAUST)
Sebastian Pazos: King Abdullah University of Science and Technology (KAUST)
Mario Lanza: King Abdullah University of Science and Technology (KAUST)

Nature Communications, 2024, vol. 15, issue 1, 1-40

Abstract: Abstract Artificial Intelligence (AI) is currently experiencing a bloom driven by deep learning (DL) techniques, which rely on networks of connected simple computing units operating in parallel. The low communication bandwidth between memory and processing units in conventional von Neumann machines does not support the requirements of emerging applications that rely extensively on large sets of data. More recent computing paradigms, such as high parallelization and near-memory computing, help alleviate the data communication bottleneck to some extent, but paradigm- shifting concepts are required. Memristors, a novel beyond-complementary metal-oxide-semiconductor (CMOS) technology, are a promising choice for memory devices due to their unique intrinsic device-level properties, enabling both storing and computing with a small, massively-parallel footprint at low power. Theoretically, this directly translates to a major boost in energy efficiency and computational throughput, but various practical challenges remain. In this work we review the latest efforts for achieving hardware-based memristive artificial neural networks (ANNs), describing with detail the working principia of each block and the different design alternatives with their own advantages and disadvantages, as well as the tools required for accurate estimation of performance metrics. Ultimately, we aim to provide a comprehensive protocol of the materials and methods involved in memristive neural networks to those aiming to start working in this field and the experts looking for a holistic approach.

Date: 2024
References: View references in EconPapers View complete reference list from CitEc
Citations: View citations in EconPapers (2)

Downloads: (external link)
https://www.nature.com/articles/s41467-024-45670-9 Abstract (text/html)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:nat:natcom:v:15:y:2024:i:1:d:10.1038_s41467-024-45670-9

Ordering information: This journal article can be ordered from
https://www.nature.com/ncomms/

DOI: 10.1038/s41467-024-45670-9

Access Statistics for this article

Nature Communications is currently edited by Nathalie Le Bot, Enda Bergin and Fiona Gillespie

More articles in Nature Communications from Nature
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().