Efficient online quantum circuit learning with no upfront training

Journal article

Optimization is a promising candidate for studying the utility of variational quantum algorithms (VQAs). However, evaluating cost functions using quantum hardware introduces runtime overheads that limit exploration. Surrogate-based methods can reduce calls to a quantum computer, yet existing approaches require hyperparameter pre-training and have been tested only on small problems. Here, we show that surrogate-based methods can enable successful optimization at scale, without pre-training, by using radial basis function interpolation (RBF) to construct an adaptive, hyperparameter-free surrogate. Using the surrogate as an acquisition function drives hardware queries to the vicinity of the true optima. For 16-qubit random 3-regular Max-Cut instances with the Quantum Approximate Optimization Algorithm (QAOA), our method outperforms state-of-the-art approaches, without considering their upfront training costs. Furthermore, we successfully optimize QAOA circuits for 127-qubit random Ising models on an IBM processor using 104−105 measurements. Strong empirical performance demonstrates the promise of automated surrogate-based learning for large-scale VQA applications.

Authors: O’Leary, T; Czarnik, P; Pelofske, E; Sornborger, AT; McKerns, M

• Publication status: Published
• Peer review status: Peer reviewed
• Publisher: Nature Research
• Journal: Communications Physics
• Volume: 8
• Issue: 1
• Article number: 514
• Publication date: 2025-11-20
• Acceptance date: 2025-11-05
• DOI: 10.1038/s42005-025-02423-4
• EISSN: 2399-3650
• ISSN: 2399-3650
• Language: English