Speakers
Description
Practical deployment of quantum algorithms on noisy intermediate-scale quantum (NISQ) devices requires error mitigation techniques that remain effective under realistic experimental constraints. Existing approaches often rely on accurate noise characterization, simplified assumptions about hardware errors, or substantial computational overhead, limiting their scalability in practical settings. In this work, we investigate two complementary strategies toward practical quantum error mitigation (QEM): exploiting experimentally accessible information about the hardware noise and learning mitigation transformations directly from measurement data.
The first approach develops a hardware-aware extension of zero-noise extrapolation (ZNE) based on structured circuit layout permutations, providing rigorous guarantees while reducing experimental overhead. The second approach introduces a data-driven QEM framework based on supervised machine learning, enabling the reconstruction of noiseless observables without requiring an explicit noise model. Together, these methods demonstrate two complementary directions for adaptive error mitigation: physics-informed protocols that exploit device characteristics and machine-learning approaches that infer mitigation strategies from data.
| Topic | Machine learning techniques for experimental quantum information science |
|---|---|
| Additional Topic | Encoding and processing of data in quantum systems |