Automating and optimizing measurement of superconducting qubits
Abstract: A major challenge in scaling quantum computation is the efficient control and calibration of many-qubit systems. Current quantum control protocols often demand substantial human oversight and are limited in detecting non-ideal terms outside the qubit Hamiltonian. Rapid identification of defects and unexpected couplings, along with robust control strategies to address them, will be crucial as systems grow in complexity. Recent advancements in autonomous characterization and control leverage machine learning and Bayesian techniques, and we aim to extend these methods into a scalable closed-loop optimization tool capable of autonomously maintaining large qubit networks. In this talk I’ll present the envisioned design of this protocol, share some initial results in optimizing tune-up of qubit parameters, and benchmark performance of three different optimization strategies.
Student Bio: Oliver Wolff is a 2nd year physics graduate student in Wolfgang Pfaff’s group studying how classical computation can augment the performance of quantum devices.