Title :
Live demonstration: Handwritten digit recognition using spiking deep belief networks on SpiNNaker
Author :
Stromatias, Evangelos ; Neil, Daniel ; Galluppi, Francesco ; Pfeiffer, Michael ; Shih-Chii Liu ; Furber, Steve
Author_Institution :
Adv. Processor Technol. Group, Univ. of Manchester, Manchester, UK
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.
Keywords :
belief networks; handwritten character recognition; image recognition; multiprocessing systems; neural chips; pattern classification; SpiNNaker chip; biologically inspired manycore architecture; classification latency; interactive handwritten digit recognition system; real-time system; spike-based deep belief network; Biological neural networks; Computer architecture; Handwriting recognition; Mobile handsets; Neurons; Real-time systems; Associated Track 8.1: Neural Networks and Systems; Spiking Neural Network circuits and systems;
Conference_Titel :
Circuits and Systems (ISCAS), 2015 IEEE International Symposium on
Conference_Location :
Lisbon
DOI :
10.1109/ISCAS.2015.7169034
