Shaking Up Quantum Simulators with Fuzzing and Rigour
summary
FuzzQ combines an Alloy-based QASM 3.0 model with structured fuzzing, mutation testing and invariant checks to generate constraint-guided circuits and run 800,000+ executions across Qiskit and Cirq. It compares gate-by-gate statevectors, measurement distributions and a costly unitary consistency check (scales to ~12–15 qubits) to spot divergences. Throughput reaches ~140 circuits/sec, depth up to ~200 gates, code coverage saturates by six qubits, and the campaign uncovers eight simulator bugs, six previously undocumented.
AI-written summary · about
chapters (9)
- 0:00Fuzzing quantum simulators
- 1:47Why rigor matters
- 3:15Framework methodology
- 4:42Differential testing
- 9:00Performance results
- 10:26Real-world bugs
- 13:16Memory efficiency trade-offs
- 15:07Fuzzing with formal methods
- 16:13Q&A
related talks
18mAn Automata-Based Framework for Verification and Bug Hunting in Quantum Circuits
19mExtracting maximum value from today's quantum computers while prep
17mQuantum Applications with Classiq
17mQVM: Quantum Gate Virtualization Machine
50mThe quantum computers are coming
53mThe quantum computers are coming
50mGod really plays dice - Introduction to Quantum Computing with Q#
01h 35mHands-On Quantum Computing