Optimized Control of Superconducting Qubits

5 views
Download
  • Share

In the past few years we could observed a rapid development of quantum technologies culminating in the realization of first quantum computer prototypes that promise to outperform conventional computers in specific types of problems. This includes problems in optimization and machine learning, but even more so in the computation of complex molecules and the simulation of condensed matter systems. This technological progress is based on our ability to control precisely the quantum state of single atoms, electrons or superconducting circuits. However, we are still facing numerous challenges that have to be tackled before practical real-life problems can be solved efficiently on quantum computers. In particular, to enhance the capabilities of today’s quantum processors not only the coherence time of the qubits, but also the preparation of complex quantum states via optimized control signals has to be further improved to reach quantum operations with low error rates. 


In this presentation I will focus on superconducting quantum circuits based on non-linear Josephson junctions as one of the most promising platforms for realizing a quantum computer. By systematically optimizing fabrication parameters to deposit and structure superconducting aluminum and niobium thin films we can realize qubits with coherence times exceeding 100 microseconds. Because of their strong coupling to microwave signals their state can be controlled within tens of nanoseconds. We then utilize closed-loop optimization methods based on high duty-cycle measurements to tailor the microwave pulses towards short and high-fidelity single-qubit gates. Using a piecewise-constant pulse parameterization we demonstrate single-qubit pulses as short as 4ns and with low leakage out of the computational subspace. We further investigate ways to reduce the number of required microwave signal lines and explore architectures that contain 'hidden' qubits. These are not directly addressable, but only over a control qubit. We first discuss the impact of such restricted control capabilities on the performance of specific qubit coupling networks. We then experimentally demonstrate full control and measurement capabilities of the hidden qubit. I will further preview promising future directions such as multi-qubit gate operations for the efficient implementation of quantum algorithms which we are developing in the newly formed Munich Quantum Valley, a cross-disciplinary initiative to realize a full-stack quantum computer.

In the past few years we could observed a rapid development of quantum technologies culminating in the realization of first quantum computer prototypes that promise to outperform conventional computers in specific types of problems. This includes problems in optimization and machine learning, but even more so in the computation of complex molecules and the simulation of condensed matter systems. This technological progress is based on our ability to control precisely the quantum state of single atoms, electrons or superconducting circuits. However, we are still facing numerous challenges that have to be tackled before practical real-life problems can be solved efficiently on quantum computers. In particular, to enhance the capabilities of today’s quantum processors not only the coherence time of the qubits, but also the preparation of complex quantum states via optimized control signals has to be further improved to reach quantum operations with low error rates. 

Advertisment

Advertisment