Coursera Flash Sale
40% Off Coursera Plus for 3 Months!
Grab it
Explore a groundbreaking approach to quantum simulator validation through this 19-minute conference presentation from OOPSLA 2025. Learn how researchers from Queen Mary University of London and King's College London developed FuzzQ, an innovative framework that combines formal methods with structured test generation and fuzzing techniques to identify critical bugs in quantum computing simulators. Discover how the team bridges the gap between existing formal validation methods and conventional testing by employing differential benchmarking, mutation testing, and invariant checking. Examine their Alloy-based formal model of QASM 3.0 that encodes quantum circuit semantics to enable automated analysis and generate structurally diverse, constraint-guided quantum circuits with guaranteed properties. Understand the novel test oracles introduced for assessing both Alloy's modeling capabilities and simulator correctness, including invariant-based checks, statistical distribution tests, and a cross-simulator unitary consistency check that reveals discrepancies missed by standard statevector comparisons. Review the comprehensive evaluation conducted on Qiskit and Cirq platforms, where over 800,000 quantum circuits were executed to assess throughput, code coverage, circuit coverage, and simulator performance metrics including sensitivity, correctness, and memory overhead. Learn about the eight simulator bugs discovered during testing, six of which were previously undocumented, and explore the outlined path for extending the framework to support mixed-state simulations under realistic noise models.
