Semi-Supervised Graph Structure Learning on Neuromorphic Computers
Authors:
Open Access
- 27 July 2022
- conference paper
- Published by Association for Computing Machinery (ACM)
- p. 1-4
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
Funding Information
- US Department of Energy (DE-AC05-00OR22725)
This publication has 9 references indexed in Scilit:
- Versatile feature learning with graph convolutions and graph structures2021
- Graph Convolutional Networks with Kalman Filtering for Traffic Prediction2020
- Provable Neuromorphic Advantages for Computing Shortest Paths2020
- Neuromorphic Graph Algorithms2020
- Graph Convolutional Networks for Text ClassificationProceedings of the AAAI Conference on Artificial Intelligence, 2019
- Shortest Path and Neighborhood Subgraph Extraction on a Spiking Memristive Neuromorphic Implementation2019
- Crystal Graph Convolutional Neural Networks for an Accurate and Interpretable Prediction of Material PropertiesPhysical Review Letters, 2018
- node2vec2016
- Collective Classification in Network DataAI Magazine, 2008