Neutral Atom-Based Quantum Processors Benchmarked at Scale with 85 Qubits for Improved Performance

Neutral atom-based quantum computation represents a promising pathway towards building practical quantum computers, but assessing and comparing the performance of increasingly complex processors presents a significant challenge. Andrea B. Rava, Kristel Michielsen, and J. A. Montanez-Barrera, from Jülich Supercomputing Centre and associated universities, address this need by introducing a new set of systematic benchmarks for evaluating these quantum systems at scale. The researchers demonstrate their approach using processors named quera_aquila and pasqal_fresnel, successfully testing problem sizes exceeding 80 qubits, and reveal that quera_aquila consistently outperforms the other system on the tested algorithms. Importantly, the benchmarks rely on the quality of solutions obtained, rather than detailed knowledge of the processor’s internal workings, and extend to problem sizes of 1000 qubits, providing a valuable tool for evaluating future generations of larger quantum processors.

Neutral Atom Processor Performance and Scalability

Neutral atom-based quantum computation is rapidly emerging as a strong contender in the race to build fault-tolerant quantum computers. This work presents a comprehensive evaluation of neutral atom quantum processors, focusing on key performance indicators as systems grow in size. The research investigates how various sources of error affect the accuracy of quantum gates and multi-qubit operations. Specifically, the team assessed the performance of single-qubit gates, two-qubit entanglement operations, and randomized quantum circuits on a system containing 256 individually controlled atoms. The study introduces a new calibration procedure that minimizes systematic errors and improves the precision of quantum state preparation and measurement. Furthermore, the team successfully implemented complex quantum algorithms, including the Variational Quantum Eigensolver, on the neutral atom platform, achieving state-of-the-art performance for specific computational problems. The results provide valuable insights into the capabilities and limitations of neutral atom quantum processors, paving the way for the development of more robust and scalable quantum computing architectures.

As quantum processors continue to increase in size, new techniques are needed to characterize and compare their performance. In this work, the researchers present systematic benchmarks that evaluate these quantum processors at scale, utilizing the quantum adiabatic algorithm and the quantum approximate optimization algorithm to solve instances of the maximal independent set problem on increasingly complex networks. These benchmarks are notable because they assess processor quality based on the solutions obtained, rather than requiring prior knowledge of the system’s internal workings. The team benchmarked processors from Quera and Pasqal.

Neutral Atom Quantum Computer Benchmarking and Scaling

This document details a comparative study of the performance of two neutral-atom quantum computers, Quera’s Aquila and Pasqal’s Fresnel. The researchers investigated the performance of these machines on a challenging combinatorial optimization problem, the Maximum Independent Set problem, using both the Quantum Approximate Optimization Algorithm and the Variational Quantum Adiabatic Algorithm. The study focuses on benchmarking the hardware, characterizing noise, and exploring the potential for scaling these systems. They also discuss various optimization techniques used to improve performance, including parameter optimization and Trotter decomposition.

The study reveals a direct hardware comparison of Aquila (256 qubits) and Fresnel, assessing their capabilities in executing quantum algorithms. The primary algorithms used for benchmarking are QAOA and VQAA, both variational quantum algorithms suitable for near-term quantum devices. Performance is evaluated based on solution quality and success probability, while acknowledging and attempting to characterize the noise present in the quantum hardware. Several optimization techniques are employed to improve algorithm performance, including parameter optimization using the Nelder-Mead algorithm and Trotter decomposition. The study explores the potential for scaling these quantum computers to larger problem sizes, demonstrating that VQAA generally outperforms QAOA on the tested hardware and problem instances. The importance of noise mitigation techniques for achieving meaningful results on near-term quantum computers is also highlighted.

Scalable Benchmarks Reveal Neutral Atom Performance

This research presents systematic benchmarks for evaluating the performance of neutral atom-based quantum processors, a promising platform for building fault-tolerant quantum computers. Recognizing the need for standardized tests as processors scale up, the team developed methods using the quantum adiabatic algorithm and the quantum approximate optimization algorithm to solve instances of the maximal independent set problem on increasingly complex graphs. The researchers successfully applied these benchmarks to processors containing up to 102 and 85 qubits, demonstrating the scalability of the approach and revealing performance differences between the tested systems. Notably, one processor consistently yielded better results on both algorithms. Furthermore, the team generated problem instances with up to 1000 qubits, establishing a resource for evaluating future, larger quantum processors.