Benchmarking neutral atom-based quantum processors at scale

TL;DR

This work uses the quantum adiabatic algorithm (QAA) and the quantum approximate optimization algorithm (QAOA) to solve maximal independent set (MIS) instances of random unit-disk graphs, providing scalable benchmarks for evaluating future, larger quantum processors as they become available.

Abstract

In recent years, neutral atom-based quantum computation has been established as a competing alternative for the realization of fault-tolerant quantum computation. However, as with other quantum technologies, various sources of noise limit their performance. With processors continuing to scale up, new techniques are needed to characterize and compare them in order to track their progress. In this work, we present two systematic benchmarks that evaluate these quantum processors at scale. We use the quantum adiabatic algorithm (QAA) and the quantum approximate optimization algorithm (QAOA) to solve maximal independent set (MIS) instances of random unit-disk graphs. These benchmarks are scalable, relying not on prior knowledge of the system's evolution but on the quality of the MIS solutions obtained. We benchmark quera_aquila and pasqal_fresnel on problem sizes up to 102 and 85 qubits, respectively. Overall, quera_aquila performs better on QAOA and QAA instances. Finally, we generate MIS instances of up to 1000 qubits, providing scalable benchmarks for evaluating future, larger processors as they become available.

Figures (5)

• Figure 1: Benchmarking of neutral atom quantum devices. (a) Arrangement of the neutral atoms for the solution of a 25-qubit instance. $a$ is the minimum distance between atoms, and $\sqrt{2}a$ is the distance where a Rydberg interaction is considered strongly enough to encode an edge of the MIS. (b) QAA protocol for a $t=4\,\mathrm{\mu s}$. (c) Valid solutions (up) and approximation ratio (down), $r$, for problem sizes between 11 and 85 qubits for quera_aquila and pasqal_fresnel. Error bars represent the standard deviation for 3 experiments of 500 samples each.
• Figure 2: Optimized schedule for the QAOA protocol.
• Figure 3: Probability distribution of different solutions comparing quera_aquila, pasqal_fresnel, and an ideal (noiseless) simulator for a problem with 13 qubits using QAA with $t=4\,\mathrm{\mu s}$.
• Figure 4: Probability of success vs. number of qubits for QAOA and QAA on quera_aquila and pasqal_fresnel. (left) $N_q\le 30$ (right) $N_q>30$.
• Figure 5: Cost of the best solution found versus number of qubits for QAA and $t=4\,\mathrm{\mu s}$. The dashed line represents the optimal cost for each instance. The error bars represent the standard deviation over the 3 experiments.