Resource frugal optimizer for quantum machine learning

Journal Article · Wed Aug 23 04:00:00 EDT 2023 · Quantum Science and Technology

DOI:https://doi.org/10.1088/2058-9565/acef55· OSTI ID:2228652

Moussa, Charles ^[1]; Hunter Gordon, Max Edward ^[2]; Baczyk, Michal ^[3]; Cerezo de la Roca, Marco Vinicio Sebastian ^[4]; Cincio, Lukasz ^[4]; Coles, Patrick J. ^[5]

---

1. Los Alamos National Laboratory (LANL), Los Alamos, NM (United States); Leiden Univ. (Netherlands)
2. Los Alamos National Laboratory (LANL), Los Alamos, NM (United States); Univ. Autonoma de Madrid (Spain)
3. Los Alamos National Laboratory (LANL), Los Alamos, NM (United States); Univ. of Warsaw (Poland)
4. Los Alamos National Laboratory (LANL), Los Alamos, NM (United States)
5. Los Alamos National Laboratory (LANL), Los Alamos, NM (United States); Normal Computing Corporation, New York, NY (United States)

Quantum-enhanced data science, also known as quantum machine learning (QML), is of growing interest as an application of near-term quantum computers. Variational QML algorithms have the potential to solve practical problems on real hardware, particularly when involving quantum data. However, training these algorithms can be challenging and calls for tailored optimization procedures. Specifically, QML applications can require a large shot-count overhead due to the large datasets involved. In this work, we advocate for simultaneous random sampling over both the dataset as well as the measurement operators that define the loss function. We consider a highly general loss function that encompasses many QML applications, and we show how to construct an unbiased estimator of its gradient. This allows us to propose a shot-frugal gradient descent optimizer called Refoqus (REsource Frugal Optimizer for QUantum Stochastic gradient descent). Our numerics indicate that Refoqus can save several orders of magnitude in shot cost, even relative to optimizers that sample over measurement operators alone.

• Research Organization: Los Alamos National Laboratory (LANL), Los Alamos, NM (United States)
• Sponsoring Organization: USDOE Laboratory Directed Research and Development (LDRD) Program; USDOE Office of Science (SC), Advanced Scientific Computing Research (ASCR); USDOE National Nuclear Security Administration (NNSA)
• Grant/Contract Number: 89233218CNA000001
• OSTI ID: 2228652
• Report Number(s): LA-UR--22-31774
• Journal Information: Quantum Science and Technology, Journal Name: Quantum Science and Technology Journal Issue: 4 Vol. 8; ISSN 2058-9565
• Publisher: IOP Publishing
• Country of Publication: United States
• Language: English

Related Subjects

97 MATHEMATICS AND COMPUTING
quantum machine learning
resource frugal
shot frugal
variational quantum algorithms