Controlling the chaotic dynamics of a monitored quantum circuit on quantum hardware
ORAL
Abstract
Programmable quantum devices offer a versatile platform for controlling the coherent dynamics of many-body wavefunctions.
Here, we study a non-unitary adaptive quantum circuit, by bringing the measurement and conditional feedback to compete with a quantum version of the classically chaotic Bernoulli map.
This adaptive quantum circuit drives the dynamics of a chaotic phase toward a controlled fixed-point state, with quantum critical point separating quantum and classical dynamics.
The transition can be characterized through quantum fluctuations of observables across different trajectories within a fixed quantum channel.
Using experimental data from IBM’s superconducting quantum processor, we benchmark the results against classical simulations based on matrix product states, as well as dephasing and depolarizing noise models.
By computing the Kullback-Leibler divergence between simulated and experimental data, we find that the experimental data aligns closely with the matrix product state simulations, indicating that the observed behavior cannot be explained by classical noise models.
Our results demonstrate the capability to steer quantum dynamics through extensive mid-circuit measurements and feedback, marking an important step toward both the state preparation and fault-tolerant quantum computing.
Here, we study a non-unitary adaptive quantum circuit, by bringing the measurement and conditional feedback to compete with a quantum version of the classically chaotic Bernoulli map.
This adaptive quantum circuit drives the dynamics of a chaotic phase toward a controlled fixed-point state, with quantum critical point separating quantum and classical dynamics.
The transition can be characterized through quantum fluctuations of observables across different trajectories within a fixed quantum channel.
Using experimental data from IBM’s superconducting quantum processor, we benchmark the results against classical simulations based on matrix product states, as well as dephasing and depolarizing noise models.
By computing the Kullback-Leibler divergence between simulated and experimental data, we find that the experimental data aligns closely with the matrix product state simulations, indicating that the observed behavior cannot be explained by classical noise models.
Our results demonstrate the capability to steer quantum dynamics through extensive mid-circuit measurements and feedback, marking an important step toward both the state preparation and fault-tolerant quantum computing.
*This work was partially supported by the US-ONR grant No. N00014-23-12357 (H.P. and J.H.P.), the Army Research Office Grant No. W911NF-23-1-0144 (K.A. and J.H.P.), NSF Grant No. DMR-2315063 (S.G.), and NSF CAREER Grants No. DMR-2143635 (T.I.), and DMR-2238895 (J.H.W.). This work was performed in part at the Aspen Center for Physics, which is supported by National Science Foundation grant PHY2210452 (T.I., J.H.W., J.H.P.) and at the Kavli Institute for Theoretical Physics (KITP), which is supported by grant NSF PHY-2309135 (J.H.P.). We are grateful to the DoD High Performance Computing Modernization Program (HPCMP) for providing us access to the Carpenter cluster.
–
Publication: https://arxiv.org/pdf/2509.18259
Presenters
-
Haining Pan
- Rutgers University