Extrapolative Quantum Error Mitigation in Continuous-Variable Systems beyond the Training Horizon

TL;DR

This paper presents a novel machine learning framework using a time-conditioned Swin Transformer to extrapolate quantum error mitigation in continuous-variable systems, enabling effective noise correction beyond the training data horizon.

Contribution

It introduces an extrapolative QEM method that leverages a time-conditioned transformer to handle long-time noise accumulation without extensive training data.

Findings

Accurately recovers quantum states under long-time noise.

Effective in both Markovian and non-Markovian noise environments.

Outperforms existing methods limited to the training horizon.

Abstract

Continuous-variable (CV) quantum systems provide a versatile platform for quantum information processing, in which quantum states can be represented in the quadrature phase space. In realistic implementations, environmental noise, primarily photon loss and dephasing, progressively degrades these states. Machine-learning-based quantum error mitigation (QEM) has recently emerged as a promising approach to suppress such noise; however, existing methods are typically limited to the training horizon and require training data that cover the entire evolution, which is experimentally demanding. Here we introduce a framework for extrapolative quantum error mitigation based on a time-conditioned Swin Transformer. By explicitly embedding the evolution time via adaptive layer normalization, the model learns a correction map that accounts for the continuous accumulation of noise while capturing nonlocal phase-space correlations. Numerical simulations under both Markovian and non-Markovian noise demonstrate accurate state recovery in the long-time regime, where existing approaches deteriorate. Our results establish extrapolative QEM as a practical route to mitigating noise in CV quantum systems without exhaustive training data.