🤖 AI Summary
In the NISQ era, anomalous quantum program behavior arises from the entanglement of software bugs and hardware noise, rendering conventional debugging methods ineffective due to their inability to disentangle these distinct root causes.
Method: We propose the first statistically grounded quantum error attribution framework, leveraging probabilistic modeling and hypothesis testing over repeated circuit executions to characterize output distributional properties and construct interpretable, quantitative discriminative metrics that separate software errors from physical noise.
Contribution/Results: Evaluated on canonical algorithms—including Grover’s, Deutsch-Jozsa, and Simon’s—the framework achieves significantly higher attribution accuracy than baseline approaches. It transcends the applicability limits of classical debugging paradigms in quantum settings and delivers the first theoretically sound and practically deployable error classification tool for quantum software engineering.
📝 Abstract
Quantum computing in the Noisy Intermediate-Scale Quantum (NISQ) era presents significant challenges in differentiating quantum software bugs from hardware noise. Traditional debugging techniques from classical software engineering cannot directly resolve this issue due to the inherently stochastic nature of quantum computation mixed with noises from NISQ computers. To address this gap, we propose a statistical approach leveraging probabilistic metrics to differentiate between quantum software bugs and hardware noise. We evaluate our methodology empirically using well-known quantum algorithms, including Grover's algorithm, Deutsch-Jozsa algorithm, and Simon's algorithm. Experimental results demonstrate the efficacy and practical applicability of our approach, providing quantum software developers with a reliable analytical tool to identify and classify unexpected behavior in quantum programs.