Real-time adaptation of quantum noise channel estimates
AI Breakdown
Get a structured breakdown of this paper — what it's about, the core idea, and key takeaways for the field.
Abstract
Estimates of noise channels for quantum gates are required for most error mitigation techniques and are desirable for informing quantum error correction decoders. These estimates can be obtained by resource-intensive off-line characterization techniques, but can become stale due to device drift and fluctuations. We propose a method to address this issue by performing real-time adaptation of noise channel estimates during the execution of a quantum algorithmic circuit using extended flag gadgets, mid-circuit measurements and Bayesian inference. We carry out analytical calculations and numerical simulations employing a Dirichlet prior distribution for the error rates in a Pauli channel to demonstrate and evaluate the technique, which can be seen as a protocol for real-time calibration of high-level gate error information.