Jose Campos

Quantum computing offers significant speedups for simulating physical, chemical, and biological systems, and for optimization and machine learning. As quantum software grows in complexity, the classical simulation of quantum computers, which has long been essential for quality assurance, becomes infeasible. This shift requires new quality-assurance methods that operate directly on real quantum computers. This paper presents the key challenges in testing large-scale quantum software and offers software engineering perspectives for addressing them.

The increasing complexity of quantum software presents significant challenges for software verification and validation, particularly in the context of unit testing. This work presents a comprehensive study on quantum-centric unit tests, comparing traditional statistical approaches with tests specifically designed for quantum circuits. These include tests that run only on a classical computer, such as the Statevector test, as well as those executable on quantum hardware, such as the Swap test and the novel Inverse test. Through an empirical study and detailed analysis on 1,796,880 mutated quantum circuits, we investigate (a) each test's ability to detect subtle discrepancies between the expected and actual states of a quantum circuit, and (b) the number of measurements required to achieve high reliability. The results demonstrate that quantum-centric tests, particularly the Statevector test and the Inverse test, provide clear advantages in terms of precision and efficiency, reducing both false positives and false negatives compared to statistical tests. This work contributes to the development of more robust and scalable strategies for testing quantum software, supporting the future adoption of fault-tolerant quantum computers and promoting more reliable practices in quantum software engineering.

Quantum software—like classic software—needs to be designed, specified, developed, and, most importantly, tested by developers. Writing tests is a complex, error-prone, and time-consuming task. Due to the particular properties of quantum physics (e.g., superposition), quantum software is inherently more complex to develop and effectively test than classical software. Nevertheless, some preliminary works have tried to bring commonly used classical testing practices for quantum computing to assess and improve the quality of quantum programs. In this chapter, we first gather 16 quantum software testing techniques that have been proposed for the IBM quantum framework, Qiskit. Then, whenever possible, we illustrate the usage of each technique (through the proposed tool that implements it, if available) on a given running example. We showcase that although several works have been proposed to ease the burn of testing quantum software, we are still in the early stages of testing in the quantum world. Researchers should focus on delivering artifacts that are usable without much hindrance to the rest of the community, and the development of quantum benchmarks should be a priority to facilitate reproducibility, replicability, and comparison between different testing techniques.

Spectrum-based Fault Localization (SBFL) uses the coverage of test cases and their outcome (pass/fail) to predict the "suspiciousness'' of program components, e.g., lines of code. SBFL is, perhaps, the most successful fault localization technique due to its simplicity and scalability. However, SBFL heuristics do not perform well in scenarios where a program may have multiple faulty components. In this work, we propose a new algorithm that "augments'' previously proposed SBFL heuristics to produce a ranked list where faulty components ranked low by base SBFL metrics are ranked significantly higher. We implement our ideas in a tool, ARTEMIS, that attempts to "bubble up'' faulty components which are ranked lower by base SBFL metrics. We compare our technique to the most popular SBFL metrics and demonstrate statistically significant improvement in the developer effort for fault localization with respect to the basic strategies.