Cover Image

Accelerated physical emulation of bayesian inference in spiking neural networks

The massively parallel nature of biological information processing plays an important role for its superiority to human-engineered computing devices. In particular, it may hold the key to overcoming the von Neumann bottleneck that limits contemporary computer architectures. Physical-model neuromorph...

Full description

Saved in:
Bibliographic Details
Main Authors: Kungl, Ákos Ferenc (Author) , Schmitt, Sebastian (Author) , Baumbach, Andreas (Author) , Dold, Dominik (Author) , Müller, Eric (Author) , Kleider, Mitja (Author) , Mauch, Christian (Author) , Breitwieser, Oliver (Author) , Leng, Luziwei (Author) , Güttler, Gilbert Maurice (Author) , Husmann, Dan (Author) , Karasenko, Vitali (Author) , Grübl, Andreas (Author) , Schemmel, Johannes (Author) , Petrovici, Mihai A. (Author)
Format: Article (Journal)
Language:English
Published: 14 November 2019
In: Frontiers in neuroscience
Year: 2019, Volume: 13
ISSN:1662-453X
DOI:10.3389/fnins.2019.01201
Online Access:Verlag, Volltext: https://doi.org/10.3389/fnins.2019.01201
Verlag: https://www.frontiersin.org/articles/10.3389/fnins.2019.01201/full
Get full text
Author Notes:Akos F. Kungl, Sebastian Schmitt, Johann Klähn, Paul Müller, Andreas Baumbach, Dominik Dold, Alexander Kugele, Eric Müller, Christoph Koke, Mitja Kleider, Christian Mauch, Oliver Breitwieser, Luziwei Leng, Nico Gürtler, Maurice Güttler, Dan Husmann, Kai Husmann, Andreas Hartel, Vitali Karasenko, Andreas Grübl, Johannes Schemmel, Karlheinz Meier and Mihai A. Petrovici
Description
Summary:The massively parallel nature of biological information processing plays an important role for its superiority to human-engineered computing devices. In particular, it may hold the key to overcoming the von Neumann bottleneck that limits contemporary computer architectures. Physical-model neuromorphic devices seek to replicate not only this inherent parallelism, but also aspects of its microscopic dynamics in analog circuits emulating neurons and synapses. However, these machines require network models that are not only adept at solving particular tasks, but that can also cope with the inherent imperfections of analog substrates. We present a spiking network model that performs Bayesian inference through sampling on the BrainScaleS neuromorphic platform, where we use it for generative and discriminative computations on visual data. By illustrating its functionality on this platform, we implicitly demonstrate its robustness to various substrate-specific distortive effects, as well as its accelerated capability for computation. These results showcase the advantages of brain-inspired physical computation and provide important building blocks for large-scale neuromorphic applications.
Item Description:Gesehen am 20.12.2019
Physical Description:Online Resource
ISSN:1662-453X
DOI:10.3389/fnins.2019.01201

Similar Items