Unsupervised learning in probabilistic neural networks with multi-state metal-oxide memristive synapses

Alexander Serb, Johannes Bill, Ali Khiat, Radu Berdan, Robert Legenstein, Themis Prodromakis

Research output: Contribution to journalArticleResearchpeer-review

Abstract

In an increasingly data-rich world the need for developing computing systems that cannot only process, but ideally also interpret big data is becoming continuously more pressing. Brain-inspired concepts have shown great promise towards addressing this need. Here we demonstrate unsupervised learning in a probabilistic neural network that utilizes metal-oxide memristive devices as multi-state synapses. Our approach can be exploited for processing unlabelled data and can adapt to time-varying clusters that underlie incoming data by supporting the capability of reversible unsupervised learning. The potential of this work is showcased through the demonstration of successful learning in the presence of corrupted input data and probabilistic neurons, thus paving the way towards robust big-data processors.
Original languageEnglish
Article number12611
JournalNature Communications
Volume7
DOIs
Publication statusPublished - 2016

Fingerprint

synapses
Unsupervised learning
Synapses
Oxides
learning
metal oxides
Metals
Learning
Neural networks
data processing equipment
Neurons
Brain
Demonstrations
pressing
neurons
brain
Equipment and Supplies
Big data

Fields of Expertise

  • Information, Communication & Computing

Cite this

Unsupervised learning in probabilistic neural networks with multi-state metal-oxide memristive synapses. / Serb, Alexander ; Bill, Johannes; Khiat, Ali; Berdan, Radu; Legenstein, Robert; Prodromakis, Themis.

In: Nature Communications , Vol. 7, 12611, 2016.

Research output: Contribution to journalArticleResearchpeer-review

Serb, Alexander ; Bill, Johannes ; Khiat, Ali ; Berdan, Radu ; Legenstein, Robert ; Prodromakis, Themis. / Unsupervised learning in probabilistic neural networks with multi-state metal-oxide memristive synapses. In: Nature Communications . 2016 ; Vol. 7.
@article{4b938efb669348c883bf3ac5f64f4655,
title = "Unsupervised learning in probabilistic neural networks with multi-state metal-oxide memristive synapses",
abstract = "In an increasingly data-rich world the need for developing computing systems that cannot only process, but ideally also interpret big data is becoming continuously more pressing. Brain-inspired concepts have shown great promise towards addressing this need. Here we demonstrate unsupervised learning in a probabilistic neural network that utilizes metal-oxide memristive devices as multi-state synapses. Our approach can be exploited for processing unlabelled data and can adapt to time-varying clusters that underlie incoming data by supporting the capability of reversible unsupervised learning. The potential of this work is showcased through the demonstration of successful learning in the presence of corrupted input data and probabilistic neurons, thus paving the way towards robust big-data processors.",
author = "Alexander Serb and Johannes Bill and Ali Khiat and Radu Berdan and Robert Legenstein and Themis Prodromakis",
year = "2016",
doi = "doi:10.1038/ncomms12611",
language = "English",
volume = "7",
journal = "Nature Communications",
issn = "2041-1723",
publisher = "Nature Publishing Group",

}

TY - JOUR

T1 - Unsupervised learning in probabilistic neural networks with multi-state metal-oxide memristive synapses

AU - Serb, Alexander

AU - Bill, Johannes

AU - Khiat, Ali

AU - Berdan, Radu

AU - Legenstein, Robert

AU - Prodromakis, Themis

PY - 2016

Y1 - 2016

N2 - In an increasingly data-rich world the need for developing computing systems that cannot only process, but ideally also interpret big data is becoming continuously more pressing. Brain-inspired concepts have shown great promise towards addressing this need. Here we demonstrate unsupervised learning in a probabilistic neural network that utilizes metal-oxide memristive devices as multi-state synapses. Our approach can be exploited for processing unlabelled data and can adapt to time-varying clusters that underlie incoming data by supporting the capability of reversible unsupervised learning. The potential of this work is showcased through the demonstration of successful learning in the presence of corrupted input data and probabilistic neurons, thus paving the way towards robust big-data processors.

AB - In an increasingly data-rich world the need for developing computing systems that cannot only process, but ideally also interpret big data is becoming continuously more pressing. Brain-inspired concepts have shown great promise towards addressing this need. Here we demonstrate unsupervised learning in a probabilistic neural network that utilizes metal-oxide memristive devices as multi-state synapses. Our approach can be exploited for processing unlabelled data and can adapt to time-varying clusters that underlie incoming data by supporting the capability of reversible unsupervised learning. The potential of this work is showcased through the demonstration of successful learning in the presence of corrupted input data and probabilistic neurons, thus paving the way towards robust big-data processors.

U2 - doi:10.1038/ncomms12611

DO - doi:10.1038/ncomms12611

M3 - Article

VL - 7

JO - Nature Communications

JF - Nature Communications

SN - 2041-1723

M1 - 12611

ER -