Tracking Non-Markovian Quantum Dynamics of a Superconducting Qubit via Deep Learning Filters
Theoretical and experimental evidence suggests harnessing quantum mechanics to execute algorithms on qubit-based quantum hardware may allow for computation exponentially more powerful than is possible with classical computers. Characterizing how qubit states evolve in time is imperative for benchmarking quantum hardware; however, this has been difficult due to the inability to fully measure a quantum state without disrupting it. A solution is weak measurement, which recent work has improved for a single qubit by leveraging the data-processing power of a recurrent neural network (RNN). The time-evolution of qubits exchanging information with their environment over long time scales is not well understood, but must be characterized to implement efficient algorithms on a quantum processor. I propose to use weak measurement and deep learning models as filters to characterize the time-evolution of a superconducting qubit exchanging information with a simplified environment. My project will contribute to the fields ongoing research in determining the time evolution of qubits and assess a technique for improving measurement accuracy of quantum systems.
Message to Sponsor
- Major: Physics
- Sponsor: Kay & Shaw Fund
- Mentor: Irfan Siddiqi