Enable Quantum Graph Neural Networks on a Single Qubits
Keywords: Quantum Machine Learning, Quantum Neural Networks, Single Qubit, NISQ Bottleneck
TL;DR: In this work, for the 1st time, we enable Quantum Graph Neural Networks with complicate graph data on a single qubit.
Abstract: In the present NISQ era, quantum machine learning encounters significant challenges in realizing practical implementations due to constrained resources. Recently, a novel single qubit methodology enabled the execution of numerous operations on a single qubit. However, the applicability of this approach to more intricate neural network architectures remains uncertain. In this study, we investigate graph data, which poses inherent challenges for machine learning due to its intricate structure. Through the utilization of quantum walk, we introduce a graph embedding approach to reduce network parameters, culminating in the design of a novel single-qubit Quantum Graph Neural Network (sQGNN) architecture. Our experimental assessments encompass both simulations and practical executions on quantum computers, revealing that the devised sQGNNs adeptly adapt to graphs of varying dimensions and compositions. Consequently, these single-qubit QGNNs make efficient use of limited qubit resources, paving the way for the realization of expansive quantum graph neural networks on resource-constrained Variational Quantum Circuits (VQCs). This breakthrough opens up new possibilities for practical applications within the current NISQ regime.
Primary Area: learning theory
Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics.
Submission Guidelines: I certify that this submission complies with the submission instructions as described on https://iclr.cc/Conferences/2024/AuthorGuide.
Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors' identity.
No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.
Submission Number: 2655
Loading