Semi-Supervised Graph Structure Learning on Neuromorphic Computers

Guojing Cong; Seung Hwan Lim; Shruti Kulkarni; Prasanna Date; Thomas Potok; Shay Snyder; Maryam Parsa; Catherine Schuman

doi:10.1145/3546790.3546821

Semi-Supervised Graph Structure Learning on Neuromorphic Computers

Guojing Cong, Seung Hwan Lim, Shruti Kulkarni, Prasanna Date, Thomas Potok, Shay Snyder, Maryam Parsa, Catherine Schuman

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

7 Scopus citations

Abstract

Graph convolutional networks have risen in popularity in recent years to tackle problems that are naturally represented as graphs. However, real-world graphs are often sparse, which means that implementing them on traditional accelerators such as graphics processing units (GPUs) can lead to inefficient utilization of the hardware. Spiking neuromorphic computers natively implement network-like computation and have been shown to be successful at implementing certain types of graph computations. In this work, we evaluate the use of a simulated network of spiking neurons to perform semi-supervised learning on graph data using only the graph structure. We demonstrate that our neuromorphic approach provides comparable results to graph convolutional network results, and we discuss the opportunities for using neuromorphic computers for this task in the future.

Original language	English
Title of host publication	ICONS 2022 - Proceedings of International Conference on Neuromorphic Systems 2022
Publisher	Association for Computing Machinery
ISBN (Electronic)	9781450397896
DOIs	https://doi.org/10.1145/3546790.3546821
State	Published - Jul 27 2022
Event	2022 International Conference on Neuromorphic Systems, ICONS 2022 - Knoxville, United States Duration: Jul 27 2022 → Jul 29 2022

Publication series

Name	ACM International Conference Proceeding Series

Conference

Conference	2022 International Conference on Neuromorphic Systems, ICONS 2022
Country/Territory	United States
City	Knoxville
Period	07/27/22 → 07/29/22

Funding

This research used resources of the Compute and Data Environment for Science (CADES) at the Oak Ridge National Laboratory, which is supported by the Office of Science of the U.S. Department of Energy under Contract No. DE-AC05-00OR22725. This material is based upon work supported by the U.S. Department of Energy, Office of Science, Office of Advanced Scientific Computing Research, under contract number DE-AC05-00OR22725.

Keywords

graph neural networks
neuromorphic computing
spike timing dependent plasticity

Access to Document

10.1145/3546790.3546821

Cite this

Cong, G., Lim, S. H., Kulkarni, S., Date, P., Potok, T., Snyder, S., Parsa, M., & Schuman, C. (2022). Semi-Supervised Graph Structure Learning on Neuromorphic Computers. In ICONS 2022 - Proceedings of International Conference on Neuromorphic Systems 2022 Article 28 (ACM International Conference Proceeding Series). Association for Computing Machinery. https://doi.org/10.1145/3546790.3546821

@inproceedings{840280c0ec804802a13f0ded92d796bc,

title = "Semi-Supervised Graph Structure Learning on Neuromorphic Computers",

abstract = "Graph convolutional networks have risen in popularity in recent years to tackle problems that are naturally represented as graphs. However, real-world graphs are often sparse, which means that implementing them on traditional accelerators such as graphics processing units (GPUs) can lead to inefficient utilization of the hardware. Spiking neuromorphic computers natively implement network-like computation and have been shown to be successful at implementing certain types of graph computations. In this work, we evaluate the use of a simulated network of spiking neurons to perform semi-supervised learning on graph data using only the graph structure. We demonstrate that our neuromorphic approach provides comparable results to graph convolutional network results, and we discuss the opportunities for using neuromorphic computers for this task in the future.",

keywords = "graph neural networks, neuromorphic computing, spike timing dependent plasticity",

author = "Guojing Cong and Lim, {Seung Hwan} and Shruti Kulkarni and Prasanna Date and Thomas Potok and Shay Snyder and Maryam Parsa and Catherine Schuman",

note = "Publisher Copyright: {\textcopyright} 2022 ACM.; 2022 International Conference on Neuromorphic Systems, ICONS 2022 ; Conference date: 27-07-2022 Through 29-07-2022",

year = "2022",

month = jul,

day = "27",

doi = "10.1145/3546790.3546821",

language = "English",

series = "ACM International Conference Proceeding Series",

publisher = "Association for Computing Machinery",

booktitle = "ICONS 2022 - Proceedings of International Conference on Neuromorphic Systems 2022",

}

Cong, G, Lim, SH , Kulkarni, S , Date, P , Potok, T, Snyder, S, Parsa, M & Schuman, C 2022, Semi-Supervised Graph Structure Learning on Neuromorphic Computers. in ICONS 2022 - Proceedings of International Conference on Neuromorphic Systems 2022., 28, ACM International Conference Proceeding Series, Association for Computing Machinery, 2022 International Conference on Neuromorphic Systems, ICONS 2022, Knoxville, United States, 07/27/22. https://doi.org/10.1145/3546790.3546821

Semi-Supervised Graph Structure Learning on Neuromorphic Computers. / Cong, Guojing; Lim, Seung Hwan ; Kulkarni, Shruti et al.
ICONS 2022 - Proceedings of International Conference on Neuromorphic Systems 2022. Association for Computing Machinery, 2022. 28 (ACM International Conference Proceeding Series).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Semi-Supervised Graph Structure Learning on Neuromorphic Computers

AU - Cong, Guojing

AU - Lim, Seung Hwan

AU - Kulkarni, Shruti

AU - Date, Prasanna

AU - Potok, Thomas

AU - Snyder, Shay

AU - Parsa, Maryam

AU - Schuman, Catherine

PY - 2022/7/27

Y1 - 2022/7/27

N2 - Graph convolutional networks have risen in popularity in recent years to tackle problems that are naturally represented as graphs. However, real-world graphs are often sparse, which means that implementing them on traditional accelerators such as graphics processing units (GPUs) can lead to inefficient utilization of the hardware. Spiking neuromorphic computers natively implement network-like computation and have been shown to be successful at implementing certain types of graph computations. In this work, we evaluate the use of a simulated network of spiking neurons to perform semi-supervised learning on graph data using only the graph structure. We demonstrate that our neuromorphic approach provides comparable results to graph convolutional network results, and we discuss the opportunities for using neuromorphic computers for this task in the future.

AB - Graph convolutional networks have risen in popularity in recent years to tackle problems that are naturally represented as graphs. However, real-world graphs are often sparse, which means that implementing them on traditional accelerators such as graphics processing units (GPUs) can lead to inefficient utilization of the hardware. Spiking neuromorphic computers natively implement network-like computation and have been shown to be successful at implementing certain types of graph computations. In this work, we evaluate the use of a simulated network of spiking neurons to perform semi-supervised learning on graph data using only the graph structure. We demonstrate that our neuromorphic approach provides comparable results to graph convolutional network results, and we discuss the opportunities for using neuromorphic computers for this task in the future.

KW - graph neural networks

KW - neuromorphic computing

KW - spike timing dependent plasticity

UR - http://www.scopus.com/inward/record.url?scp=85138349800&partnerID=8YFLogxK

U2 - 10.1145/3546790.3546821

DO - 10.1145/3546790.3546821

M3 - Conference contribution

AN - SCOPUS:85138349800

T3 - ACM International Conference Proceeding Series

BT - ICONS 2022 - Proceedings of International Conference on Neuromorphic Systems 2022

PB - Association for Computing Machinery

T2 - 2022 International Conference on Neuromorphic Systems, ICONS 2022

Y2 - 27 July 2022 through 29 July 2022

ER -

Semi-Supervised Graph Structure Learning on Neuromorphic Computers

Abstract

Publication series

Conference

Funding

Keywords

Access to Document

Other files and links

Fingerprint

Cite this