Header

UZH-Logo

Maintenance Infos

Practical gradient-descent for memristive crossbars


Nair, Manu V; Dudek, Piotr (2015). Practical gradient-descent for memristive crossbars. In: 2015 International Conference on Memristive Systems (MEMRISYS), Paphos, Cyprus, 8 November 2015 - 10 November 2015.

Abstract

This paper discusses implementations of gradientdescent based learning algorithms on memristive crossbar arrays. The Unregulated Step Descent (USD) is described as a practical algorithm for feed-forward on-line training of large crossbar arrays. It allows fast feed-forward fully parallel on-line hardware based learning, without requiring accurate models of the memristor behaviour and precise control of the programming pulses. The effect of device parameters, training parameters, and device variability on the learning performance of crossbar arrays trained using the USD algorithm has been studied via simulations.

Abstract

This paper discusses implementations of gradientdescent based learning algorithms on memristive crossbar arrays. The Unregulated Step Descent (USD) is described as a practical algorithm for feed-forward on-line training of large crossbar arrays. It allows fast feed-forward fully parallel on-line hardware based learning, without requiring accurate models of the memristor behaviour and precise control of the programming pulses. The effect of device parameters, training parameters, and device variability on the learning performance of crossbar arrays trained using the USD algorithm has been studied via simulations.

Statistics

Altmetrics

Downloads

7 downloads since deposited on 27 Jan 2017
7 downloads since 12 months
Detailed statistics

Additional indexing

Item Type:Conference or Workshop Item (Speech), refereed, original work
Communities & Collections:07 Faculty of Science > Institute of Neuroinformatics
Dewey Decimal Classification:570 Life sciences; biology
Language:English
Event End Date:10 November 2015
Deposited On:27 Jan 2017 08:31
Last Modified:01 Sep 2017 07:16
Publisher:2015 International Conference on Memristive Systems (MEMRISYS)
Series Name:2015 International Conference on Memristive Systems (MEMRISYS)
Publisher DOI:https://doi.org/10.1109/MEMRISYS.2015.7378392

Download

Preview Icon on Download
Preview
Filetype: PDF
Size: 259kB
View at publisher