Rapid Safety Diagnosis of Regional Buildings Using Hybrid Quantum Machine Learning

Abstract

Seismic damage evaluations on buildings are often performed by government or local organizations mobilizing the experts to earthquake-affected areas, which can be time-consuming, labor-intensive, and risky. In recent years, numerous research has been conducted using machine learning and deep learning techniques, to assess the damage to building structures after an earthquake. However, the use of quantum-enhanced machine learning (QML) has a fewer work on damage assessment, which has advantages over classical machine learning (CML) algorithms in terms of larger datasets, computationally time and prediction accuracy as suggested by recent studies on various domain. Thus, this study currently examines the feasibility of using QML for rapid assessments of RC building safety after an earthquake in terms of classification accuracy. Furthermore, the performance of QML on testing and validation datasets is compared with the outcomes of widely used CML algorithms.