Binary Associative Memories as a Benchmark for Spiking Neuromorphic Hardware

Stöckel A, Jenzen C, Thies M, Rückert U (2017)
Frontiers in Computational Neuroscience 11: 71.

Download
OA 1.13 MB
Zeitschriftenaufsatz | Veröffentlicht | Englisch
Abstract / Bemerkung
Large-scale neuromorphic hardware platforms, specialized computer systems for energy efficient simulation of spiking neural networks, are being developed around the world, for example as part of the European Human Brain Project (HBP). Due to conceptual differences, a universal performance analysis of these systems in terms of runtime, accuracy and energy efficiency is non-trivial, yet indispensable for further hard- and software development. In this paper we describe a scalable benchmark based on a spiking neural network implementation of the binary neural associative memory. We treat neuromorphic hardware and software simulators as black-boxes and execute exactly the same network description across all devices. Experiments on the HBP platforms under varying configurations of the associative memory show that the presented method allows to test the quality of the neuron model implementation, and to explain significant deviations from the expected reference output.
Erscheinungsjahr
Zeitschriftentitel
Frontiers in Computational Neuroscience
Band
11
Artikelnummer
71
ISSN
Finanzierungs-Informationen
Article Processing Charge funded by the Deutsche Forschungsgemeinschaft and the Open Access Publication Fund of Bielefeld University.
PUB-ID

Zitieren

Stöckel A, Jenzen C, Thies M, Rückert U. Binary Associative Memories as a Benchmark for Spiking Neuromorphic Hardware. Frontiers in Computational Neuroscience. 2017;11: 71.
Stöckel, A., Jenzen, C., Thies, M., & Rückert, U. (2017). Binary Associative Memories as a Benchmark for Spiking Neuromorphic Hardware. Frontiers in Computational Neuroscience, 11, 71. doi:10.3389/fncom.2017.00071
Stöckel, A., Jenzen, C., Thies, M., and Rückert, U. (2017). Binary Associative Memories as a Benchmark for Spiking Neuromorphic Hardware. Frontiers in Computational Neuroscience 11:71.
Stöckel, A., et al., 2017. Binary Associative Memories as a Benchmark for Spiking Neuromorphic Hardware. Frontiers in Computational Neuroscience, 11: 71.
A. Stöckel, et al., “Binary Associative Memories as a Benchmark for Spiking Neuromorphic Hardware”, Frontiers in Computational Neuroscience, vol. 11, 2017, : 71.
Stöckel, A., Jenzen, C., Thies, M., Rückert, U.: Binary Associative Memories as a Benchmark for Spiking Neuromorphic Hardware. Frontiers in Computational Neuroscience. 11, : 71 (2017).
Stöckel, Andreas, Jenzen, Christoph, Thies, Michael, and Rückert, Ulrich. “Binary Associative Memories as a Benchmark for Spiking Neuromorphic Hardware”. Frontiers in Computational Neuroscience 11 (2017): 71.
Alle Dateien verfügbar unter der/den folgenden Lizenz(en):
Copyright Statement:
This Item is protected by copyright and/or related rights. [...]
Volltext(e)
Access Level
OA Open Access
Zuletzt Hochgeladen
2018-01-15T09:02:10Z

54 References

Daten bereitgestellt von Europe PubMed Central.


Traub R., Miles R.., 1991
Full-scale simulation of a cortical microcircuit on SpiNNaker
Van S., Rowley A., Hopkins M., Schmidt M., Senk J., Stokes A.., 2016
Non-holographic associative memory.
Willshaw DJ, Buneman OP, Longuet-Higgins HC., Nature 222(5197), 1969
PMID: 5789326
GeNN: a code generation framework for accelerated brain simulations.
Yavuz E, Turner J, Nowotny T., Sci Rep 6(), 2016
PMID: 26740369

Export

Markieren/ Markierung löschen
Markierte Publikationen

Open Data PUB

Web of Science

Dieser Datensatz im Web of Science®

Quellen

PMID: 28878642
PubMed | Europe PMC

Suchen in

Google Scholar