Live Demonstration: Handwritten Digit Recognition Using Spiking Deep Belief Networks on SpiNNaker

Evangelos Stromatias, Daniel Neil, Francesco Galluppi, Michael Pfeiffer, Shih-Chii Liu, Steve Furber

Research output: Contribution to conferencePoster

346 Downloads (Pure)

Abstract

We demonstrate an interactive handwritten digit recognition system with a spike-based deep belief network running in real-time on SpiNNaker, a biologically inspired many-core architecture. Results show that during the simulation a SpiNNaker chip can deliver spikes in under 1 μs, with a classification latency in the order of tens of milliseconds, while consuming less than 0.3 W.
Original languageEnglish
DOIs
Publication statusPublished - 2015
Event2015 IEEE International Symposium on Circuits & Systems - Lisbon, portugal
Duration: 24 May 201527 May 2015
http://www.iscas2015.org/

Conference

Conference2015 IEEE International Symposium on Circuits & Systems
CityLisbon, portugal
Period24/05/1527/05/15
Internet address

Fingerprint

Dive into the research topics of 'Live Demonstration: Handwritten Digit Recognition Using Spiking Deep Belief Networks on SpiNNaker'. Together they form a unique fingerprint.

Cite this