Quantum non-Markovian response spectra
Abstract
Quantum systems are rarely isolated from their surroundings. These hidden environments can retain information and return it later, producing correlated errors in quantum computers and complicating the interpretation of experiments. On a quantum processor, for example, one job can therefore leave an environmental trace that affects the next, even after the qubits themselves have been reset. Learning how this hidden influence persists and affects later behavior is therefore important, but the environment is often impossible to measure directly, and fully reconstructing its effects through tomography rapidly becomes impractical. Here, we introduce a response matrix that uses controlled interventions and observations of the system alone to reveal the temporal structure of environmental memory. The matrix records how future observations change across different controlled histories. Its spectrum characterizes how memory is structured and how it persists under specified system resets, while suitable linear combinations of its entries can rule out specified classical models of that memory. Because its history-future queries can be evaluated independently, numerical response matrices can be assembled in parallel. We establish the framework theoretically, characterize response spectra in an interacting spin process, and certify quantum memory in a two-qubit benchmark. This approach turns system observations into a probe of inaccessible environments and could support the diagnosis of correlated noise, tests of open system models, and memory-aware control design.
Related
- The Memory Hidden in Response Fluctuations: Trajectory-Level Fluctuation-Response Theory and Inequalities for Non-Markovian Jump Dynamics
- Non-Markovian delay-assisted sensing with waveguide-coupled quantum emitters
- Quantum Simulation of Markovian and Non-Markovian Open Quantum Dynamics in Heavy-Ion Collisions
- Open-quantum-system theory of non-Markovian electron-phonon dynamics
- Non-Markovian interactions with pulses of quantum radiation
BibTeX
@misc{sander2026quantum,
title = {{Quantum non-Markovian response spectra}},
author = {Aaron Sander and John F. Kam and Gauthameshwar S. and Robert Wille and Kavan Modi},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.10684},
eprint = {2610.10684},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.10684},
}
RIS
TY - UNPB
TI - Quantum non-Markovian response spectra
AU - Aaron Sander
AU - John F. Kam
AU - Gauthameshwar S.
AU - Robert Wille
AU - Kavan Modi
PY - 2026
DA - 2026/10/09
PB - arXiv
DO - 10.48550/arxiv.2610.10684
UR - https://arxiv.org/abs/2610.10684
AB - Quantum systems are rarely isolated from their surroundings. These hidden environments can retain information and return it later, producing correlated errors in quantum computers and complicating the interpretation of experiments. On a quantum processor, for example, one job can therefore leave an environmental trace that affects the next, even after the qubits themselves have been reset. Learning how this hidden influence persists and affects later behavior is therefore important, but the environment is often impossible to measure directly, and fully reconstructing its effects through tomography rapidly becomes impractical. Here, we introduce a response matrix that uses controlled interventions and observations of the system alone to reveal the temporal structure of environmental memory. The matrix records how future observations change across different controlled histories. Its spectrum characterizes how memory is structured and how it persists under specified system resets, while suitable linear combinations of its entries can rule out specified classical models of that memory. Because its history-future queries can be evaluated independently, numerical response matrices can be assembled in parallel. We establish the framework theoretically, characterize response spectra in an interacting spin process, and certify quantum memory in a two-qubit benchmark. This approach turns system observations into a probe of inaccessible environments and could support the diagnosis of correlated noise, tests of open system models, and memory-aware control design.
ER -