🤖 AI Summary
On current NISQ devices, accurately estimating and comparing the noise performance of quantum circuit layouts remains challenging due to substantial variability in gate error rates and the absence of a standardized, hardware-agnostic fidelity metric. To address this, we propose Q-fid—a novel system that frames fidelity prediction as a time-series causal modeling problem over gate sequences, leveraging an LSTM network to learn hardware-aware, dynamic error propagation patterns. Our key contributions are: (1) the first LSTM-based fidelity prediction framework tailored for quantum circuits; (2) a standardized, comparable circuit-level fidelity metric grounded in realistic noise models; and (3) cross-platform dynamic self-calibration capability. Experimental evaluation demonstrates that Q-fid reduces average RMSE significantly and improves fidelity prediction accuracy by 27% over Qiskit’s transpile, enabling more effective noise-resilient compilation and layout optimization.
📝 Abstract
The fidelity of quantum circuits (QC) is influenced by several factors, including hardware characteristics, calibration status, and the transpilation process, all of which impact their susceptibility to noise. However, existing methods struggle to estimate and compare the noise performance of different circuit layouts due to fluctuating error rates and the absence of a standardized fidelity metric. In this work, Q‐fid is introduced, a Long Short‐Term Memory (LSTM) based fidelity prediction system accompanied by a novel metric designed to quantify the fidelity of quantum circuits. Q‐fid provides an intuitive way to predict the noise performance of Noisy Intermediate‐Scale Quantum (NISQ) circuits. This approach frames fidelity prediction as a Time Series Forecasting problem to analyze the tokenized circuits, capturing the causal dependence of the gate sequences and their impact on overall fidelity. Additionally, the model is capable of dynamically adapting to changes in hardware characteristics, ensuring accurate fidelity predictions under varying conditions. Q‐fid achieves a high prediction accuracy with an average RMSE of , up to more accurate than the Qiskit transpile tool mapomatic. By offering a reliable method for fidelity prediction, Q‐fid empowers developers to optimize transpilation strategies, leading to more efficient and noise‐resilient quantum circuit implementations.
