Practical gradient-descent for memristive crossbars

Conference object, Other literature type English OPEN

Nair, Manu V; Dudek, Piotr;

(2015)

Publisher: 2015 International Conference on Memristive Systems (MEMRISYS)

Identifiers: doi: 10.1109/MEMRISYS.2015.7378392, doi: 10.5167/uzh-132662
Subject: Institute of Neuroinformatics | 570 Life sciences; biology

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-forw... View more

Metrics

Share - Bookmark

Download from
Zurich Open Repository and Archive via Zurich Open Repository and Archive (Conference object, 2015)
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. via Datacite (Other literature type, 2015)
2015 International Conference on Memristive Systems (MEMRISYS)
Cite this publication