Quantum graph neural networks compete with classical models using fewer parameters
Quantum Graph Convolutional Networks: Implementation and Trainability Analysis
Machine Learning
Summary
Graph neural networks help computers learn from network-like data, but they often run into memory problems with large graphs. This paper looks at quantum computing as a new way to handle these problems by using quantum versions of these networks. The authors tested two quantum models on common graph tasks and found they perform about as well as classical methods with fewer parameters. They also studied when these quantum models can be trained effectively and how stable they remain during training.
What this means in practice
- •For quantum computing engineers: Implement quantum graph neural networks with fewer parameters for scalable graph data learning on quantum devices.
- •For machine learning system developers: Integrate quantum-inspired graph learning techniques to optimize resource use when training models on large graph datasets.
Authors
Paul San Sebastian Sein, Theodor Iosif, Tilen G. Limbäck-Stokin, Kin Ian Lo, Yidong Liao
Abstract
Graph Neural Networks (GNNs) achieve state-of-the-art performance on graph-structured data, but training and inference on large graphs are often bottlenecked by memory constraints and sparse linear-algebra workloads. Quantum computing offers an alternative set of primitives that may improve scalability for graph learning. Building on the quantum graph neural network (QGNN) framework of Liao \textit{et al.}, this work implements two representative architectures --- the Simplified Graph Convolution (SGC) and Linear Graph Convolution (LGC) models --- and evaluates them on open benchmark graph datasets and semi-supervised learning tasks using quantum simulation. We compare predictive performance and optimization behavior against classical baselines, showing that the quantum models achieve competitive performance with fewer parameters. Finally, we present a cost gradient analysis that identifies the tasks for which the models showcased are trainable. This is followed by a classical simulability study to find regimes in which the proposed circuits remain robust during training.