PHD SEMINAR 2023 | Enhancing Qubit Readout with Autoencoders

The realization of a computer that exploits quantum – rather than classical – principles represents a formidable scientific and technological challenge. Today, superconducting quantum processors are achieving significant results in simulation and computation capabilities. However, the realization of a fault-tolerant quantum device still poses many technical difficulties. Among the many, the ability to perform high fidelity qubit readout, to actually extract information from the device, is of primary importance. In the dispersive readout technique, the qubit is coupled to a readout resonator and its state is inferred by measuring the quadrature amplitudes of an electromagnetic field transmitted through the resonator. However, random thermal noise in the hardware, gate errors or qubit decay processes occurring during the measurements can reduce the readout fidelity. Appropriate statistical models or more complex machine learning techniques can help to restore good fidelity.  A novel semi-unsupervised machine learning classification method based on autoencoder pre-training is presented and compared with state-of-the-art techniques.