A Spiking Neural Algorithm for Network Flow
Keywords
Loading...
Authors
Issue Date
2019-06-05
Language
en
Document type
Journal Title
Journal ISSN
Volume Title
Publisher
Title
ISSN
Volume
Issue
Startpage
Endpage
DOI
Abstract
It is not clear what the potential is of neuromorphic hardware beyond machine learning and neuroscience.
In this project, a problem is investigated that is inherently difficult to fully implement
in neuromorphic hardware by introducing a new machine model in which a conventional Turing machine
and neuromorphic oracle work together to solve such types of problems. A lattice of complexity
classes is introduced: CSNN(RS), in which a neuromorphic oracle is consulted using only resources at
most RS. We show that the P-complete MAX NETWORK FLOW problem is in LSNN(O(n);O(n);O(n))
for graphs with n edges. A modified variant of this algorithm is implemented on the Intel Loihi chip;
a neuromorphic manycore processor developed by Intel Labs. We show that by off-loading the search
for augmenting paths to the neuromorphic processor we can get energy efficiency gains, while not
sacrificing runtime resources. This result demonstrates how P-complete problems can be mapped on
neuromorphic architectures in a theoretically and potentially practically efficient manner.
Description
Citation
Supervisor
Faculty
Faculteit der Sociale Wetenschappen