Title:	WACQTQuantumComputerCalibration_AI4Science
DNr:	Berzelius-2026-313
Project Type:	LiU Berzelius
Principal Investigator:	Göran Johansson <goran.l.johansson@chalmers.se>
Affiliation:	Chalmers tekniska högskola
Duration:	2026-11-01 – 2027-05-01
Classification:	10304
Keywords:

Abstract

This project is done in collaboration with AI4S The project aims to establish proof-of-concept AI solutions and reusable methodological blueprints that the group can subsequently extend across the full calibration chain. The specific aims are: • Develop AI-assisted quality assurance for automated calibration of superconducting quantum processors through automatic failure detection, diagnosis, and recalibration scheduling, establishing a diagnostic pattern that can be replicated systematically across calibration nodes. • Develop machine-learning-based analysis methods for two-dimensional calibration maps to improve the robustness and accuracy of superconducting qubit gate calibration, providing a reusable CNN-based analysis pattern applicable to the growing family of 2D maps in the calibration chain. • Reduce quantum processor bring-up time by developing predictive models that identify and prioritize the most promising two-qubit gate operating points. • Quantify and model long-term device stability using time-series analysis of calibration records to establish data-driven recalibration schedules and improve processor reliability.