QuantumDARTS: Differentiable Quantum Architecture Search for Variational Quantum Algorithms

Wenjie Wu, Ge Yan, Xudong Lu, Kaisen Pan, Junchi Yan
Proceedings of the 40th International Conference on Machine Learning, PMLR 202:37745-37764, 2023.

Abstract

With the arrival of the Noisy Intermediate-Scale Quantum (NISQ) era and the fast development of machine learning, variational quantum algorithms (VQA) including Variational Quantum Eigensolver (VQE) and quantum neural network (QNN) have received increasing attention with wide potential applications in foreseeable near future. We study the problem of quantum architecture search (QAS) for VQA to automatically design parameterized quantum circuits (PQC). We devise a differentiable searching algorithm based on Gumbel-Softmax in contrast to peer methods that often require numerous circuit sampling and evaluation. Two versions of our algorithm are provided, namely macro search and micro search, where macro search directly searches for the whole circuit like other literature while the innovative micro search is able to infer the sub-circuit structure from a small-scale and then transfer that to a large-scale problem. We conduct intensive experiments on unweighted Max-Cut, ground state energy estimation, and image classification. The superior performance shows the efficiency and capability of macro search, which requires little prior knowledge. Moreover, the experiments on micro search show the potential of our algorithm for large-scale QAS problems.

Cite this Paper

BibTeX
@InProceedings{pmlr-v202-wu23v, title = {{Q}uantum{DARTS}: Differentiable Quantum Architecture Search for Variational Quantum Algorithms}, author = {Wu, Wenjie and Yan, Ge and Lu, Xudong and Pan, Kaisen and Yan, Junchi}, booktitle = {Proceedings of the 40th International Conference on Machine Learning}, pages = {37745--37764}, year = {2023}, editor = {Krause, Andreas and Brunskill, Emma and Cho, Kyunghyun and Engelhardt, Barbara and Sabato, Sivan and Scarlett, Jonathan}, volume = {202}, series = {Proceedings of Machine Learning Research}, month = {23--29 Jul}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v202/wu23v/wu23v.pdf}, url = {https://proceedings.mlr.press/v202/wu23v.html}, abstract = {With the arrival of the Noisy Intermediate-Scale Quantum (NISQ) era and the fast development of machine learning, variational quantum algorithms (VQA) including Variational Quantum Eigensolver (VQE) and quantum neural network (QNN) have received increasing attention with wide potential applications in foreseeable near future. We study the problem of quantum architecture search (QAS) for VQA to automatically design parameterized quantum circuits (PQC). We devise a differentiable searching algorithm based on Gumbel-Softmax in contrast to peer methods that often require numerous circuit sampling and evaluation. Two versions of our algorithm are provided, namely macro search and micro search, where macro search directly searches for the whole circuit like other literature while the innovative micro search is able to infer the sub-circuit structure from a small-scale and then transfer that to a large-scale problem. We conduct intensive experiments on unweighted Max-Cut, ground state energy estimation, and image classification. The superior performance shows the efficiency and capability of macro search, which requires little prior knowledge. Moreover, the experiments on micro search show the potential of our algorithm for large-scale QAS problems.} }
Endnote
%0 Conference Paper %T QuantumDARTS: Differentiable Quantum Architecture Search for Variational Quantum Algorithms %A Wenjie Wu %A Ge Yan %A Xudong Lu %A Kaisen Pan %A Junchi Yan %B Proceedings of the 40th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2023 %E Andreas Krause %E Emma Brunskill %E Kyunghyun Cho %E Barbara Engelhardt %E Sivan Sabato %E Jonathan Scarlett %F pmlr-v202-wu23v %I PMLR %P 37745--37764 %U https://proceedings.mlr.press/v202/wu23v.html %V 202 %X With the arrival of the Noisy Intermediate-Scale Quantum (NISQ) era and the fast development of machine learning, variational quantum algorithms (VQA) including Variational Quantum Eigensolver (VQE) and quantum neural network (QNN) have received increasing attention with wide potential applications in foreseeable near future. We study the problem of quantum architecture search (QAS) for VQA to automatically design parameterized quantum circuits (PQC). We devise a differentiable searching algorithm based on Gumbel-Softmax in contrast to peer methods that often require numerous circuit sampling and evaluation. Two versions of our algorithm are provided, namely macro search and micro search, where macro search directly searches for the whole circuit like other literature while the innovative micro search is able to infer the sub-circuit structure from a small-scale and then transfer that to a large-scale problem. We conduct intensive experiments on unweighted Max-Cut, ground state energy estimation, and image classification. The superior performance shows the efficiency and capability of macro search, which requires little prior knowledge. Moreover, the experiments on micro search show the potential of our algorithm for large-scale QAS problems.
APA
Wu, W., Yan, G., Lu, X., Pan, K. & Yan, J.. (2023). QuantumDARTS: Differentiable Quantum Architecture Search for Variational Quantum Algorithms. Proceedings of the 40th International Conference on Machine Learning, in Proceedings of Machine Learning Research 202:37745-37764 Available from https://proceedings.mlr.press/v202/wu23v.html.

Related Material