A quantum computer is like a finicky musical instrument: the slightest change in temperature, humidity, or string tuning, and the melody sounds different. In hybrid algorithms, where a quantum chip works alongside a classical computer, there are dozens of such 'strings': the choice of quantum circuit, error suppression method, number of shots, and even the software version. If you don't record every last detail, reproducing the 'melody'—that is, getting the same result — is nearly impossible.
The authors propose a solution: an ontology — a rigorous yet flexible schema for describing all experimental conditions. It acts like a detailed checklist for an orchestra: every item (circuit type, noise model, correction method) must be filled in, or the program highlights the omission. With special queries, you can quickly find, for instance, all runs using a particular optimizer or library version. This turns the chaos of code, logs, and configurations into a readable digital trail.
This approach makes quantum computing more transparent. If reproducing someone else's result used to be like assembling a puzzle from scattered pieces, now all the pieces are in one box with labeled compartments. It not only saves time but also bolsters confidence in the findings — after all, every note in the complex quantum symphony is now documented.
as written by the authors arXiv:2607.03982
Variational quantum algorithms are hybrid quantum-classical workflows whose results depend on many interacting choices, including the ansatz, Hamiltonian, optimizer, backend, shot count, noise model, mitigation method, random seed, stopping criteria, and software versions. In current practice, this information is often scattered across code, configuration files, logs, backend metadata, and paper descriptions, making executions difficult to reproduce, compare, debug, and reuse. This paper proposes an ontology-supported framework for representing Variational Quantum Algorithm (VQA) execution records as structured and machine-readable software engineering artifacts. The framework defines a Web Ontology Language (OWL) ontology for modeling the main entities involved in VQA experimentation, including algorithms, circuits, ansatzes, Hamiltonians, optimizers, backends, noise models, mitigation techniques, execution steps, software environments, measurement outcomes, and results. It further combines the ontology with Shapes Constraint Language (SHACL) constraints for validating completeness and consistency, and SPARQL Protocol and RDF Query Language (SPARQL) competency queries for retrieving reproducibility-relevant information. We demonstrate the approach using Variational Quantum Eigensolver (VQE) execution records, including a valid record and intentionally incomplete or inconsistent examples. The results show that the framework can represent complete VQA execution contexts, detect missing or malformed metadata, and support query-based inspection of information needed for reproducible quantum software experimentation.
Ontologies like the one proposed have long been used in biology to classify species or genes, but this is the first time such an approach has been applied to quantum computing.