Quantum Information Science: Quantum computing, quantum machine learning, and quantum materials

Speaker: Dr. Zongrui Pei
Affiliation: University of Texas at Arlington
Date: November 13, 2026

Abstract:

We will first briefly discuss the basic concepts and principles underlying quantum information science, lay the groundwork for our deeper discussion of quantum computing and quantum machine learning (QML), and eventually discuss one of its possible physical realizations with quantum materials. This talk focuses on QML algorithm design and applications. QML offers new opportunities to address complex scientific problems by exploiting quantum-state representations, entanglement, and hybrid quantum-classical algorithms. Here, I present our recent developments in QML, spanning materials discovery, property prediction, and quantum natural language processing. First, we develop quantum support vector machines (QSVMs) with optimized quantum kernels to predict phase formation in complex concentrated alloys, achieving approximately 90% classification accuracy. Combined with language modeling, this approach identifies 1,741 candidate lightweight alloys. Second, we investigate QSVMs and hybrid quantum neural networks to predict stacking fault energies and identify solutes that improve magnesium ductility, achieving validation scores of about 90% with optimized quantum circuits. Third, we systematically investigate quantum language models, examining how circuit ansätze, expressivity, and hyperparameters influence text-classification performance. Across these applications, we examine how quantum feature encoding, entanglement, circuit architecture, and noise shape model performance. Our results demonstrate the potential of QML for scientific applications while highlighting challenges in developing efficient, scalable, and reliable quantum algorithms.