Setting angles in quantum approximate optimization at utility-scale
Abstract
The quantum approximate optimization algorithm (QAOA) is a powerful heuristic that seeks to solve combinatorial optimization problems using quantum hardware and classical optimization in tandem. Various methods exist to train the parameterized quantum circuits that serve as an ansatz in QAOA. However, which method works best to identify optimal angles for a given problem instance remains poorly understood, especially at utility-scale, i.e., $100$ qubits or more. In this work, we address this challenge through utility-scale benchmarks from which we distill operational guidance for QAOA practitioners. First, we investigate approximation techniques, such as matrix product states and Pauli propagation, to find optimal angles. Second, we train QAOA on small-scale representative problems and transfer the angles to larger ones. We then validate the results on quantum hardware for utility-scale problem instances that can be meaningfully executed. In this way, we identify insights for QAOA angle setting strategies that work best for problems at the utility scale, including as a function of resource cost for the search. Crucially, the operational implications we draw from our benchmarks will help quantum optimization practitioners execute QAOA end-to-end pipelines efficiently on current and future hardware.
Related
- Learning angles for quantum approximate optimization
- Binary Optimization with Complex Constraints via Quantum Approximate Multi-Objective Optimization
- Depth analysis of the Quantum Approximate Optimization Algorithm with a Grover mixer
- Optimal FALQON for Quantum Approximate Optimization via Layer-wise Parameter Tuning
- The Overlap Gap Property: Separating Quantum and Quantum-Inspired Approximate Optimization Algorithms
BibTeX
@misc{guo2026setting,
title = {{Setting angles in quantum approximate optimization at utility-scale}},
author = {Maosheng Guo and Joel Jurado Diaz and Anurag Ramesh and Conrad J. Haupt and Alberto Baiardi and Dimitrios Athanasakos and M. Emre Sahin and Oscar Wallis and George Pennington and Christian Arenz and Sebastian Brandhofer and Georgios Korpas and Ieva Čepaitė and J. A. Montañez-Barrera and Jakub Marecek and Davide Venturelli and Stephan Eidenbenz and David E. Bernal Neira and Daniel J. Egger},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2606.05311},
eprint = {2606.05311},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2606.05311},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Setting angles in quantum approximate optimization at utility-scale
AU - Maosheng Guo
AU - Joel Jurado Diaz
AU - Anurag Ramesh
AU - Conrad J. Haupt
AU - Alberto Baiardi
AU - Dimitrios Athanasakos
AU - M. Emre Sahin
AU - Oscar Wallis
AU - George Pennington
AU - Christian Arenz
AU - Sebastian Brandhofer
AU - Georgios Korpas
AU - Ieva Čepaitė
AU - J. A. Montañez-Barrera
AU - Jakub Marecek
AU - Davide Venturelli
AU - Stephan Eidenbenz
AU - David E. Bernal Neira
AU - Daniel J. Egger
PY - 2026
DA - 2026/09/23
PB - arXiv
DO - 10.48550/arxiv.2606.05311
UR - https://arxiv.org/abs/2606.05311
AB - The quantum approximate optimization algorithm (QAOA) is a powerful heuristic that seeks to solve combinatorial optimization problems using quantum hardware and classical optimization in tandem. Various methods exist to train the parameterized quantum circuits that serve as an ansatz in QAOA. However, which method works best to identify optimal angles for a given problem instance remains poorly understood, especially at utility-scale, i.e., $100$ qubits or more. In this work, we address this challenge through utility-scale benchmarks from which we distill operational guidance for QAOA practitioners. First, we investigate approximation techniques, such as matrix product states and Pauli propagation, to find optimal angles. Second, we train QAOA on small-scale representative problems and transfer the angles to larger ones. We then validate the results on quantum hardware for utility-scale problem instances that can be meaningfully executed. In this way, we identify insights for QAOA angle setting strategies that work best for problems at the utility scale, including as a function of resource cost for the search. Crucially, the operational implications we draw from our benchmarks will help quantum optimization practitioners execute QAOA end-to-end pipelines efficiently on current and future hardware.
ER -