[Paper Review] Quantum-centric Supercomputing for Materials Science: A Perspective on Challenges and Future Directions

Yuri Alexeev, Maximilian Amsler+125 more|arXiv (Cornell University)|Dec 14, 2023

Machine Learning in Materials Science467 references4 citations

TL;DR

This paper proposes a quantum-centric supercomputing framework integrating quantum computing, high-performance computing (HPC), and machine learning to accelerate materials science simulations. It addresses challenges in scaling noisy intermediate-scale quantum (NISQ) devices and fault-tolerant architectures, emphasizing hybrid quantum-classical workloads and continuous variable optimization, with the key contribution being a roadmap for achieving quantum advantage in materials design through synergistic quantum-classical HPC integration.

ABSTRACT

Computational models are an essential tool for the design, characterization, and discovery of novel materials. Hard computational tasks in materials science stretch the limits of existing high-performance supercomputing centers, consuming much of their simulation, analysis, and data resources. Quantum computing, on the other hand, is an emerging technology with the potential to accelerate many of the computational tasks needed for materials science. In order to do that, the quantum technology must interact with conventional high-performance computing in several ways: approximate results validation, identification of hard problems, and synergies in quantum-centric supercomputing. In this paper, we provide a perspective on how quantum-centric supercomputing can help address critical computational problems in materials science, the challenges to face in order to solve representative use cases, and new suggested directions.

Motivation & Objective

• To identify and define the challenges in achieving quantum advantage for materials science using quantum-centric supercomputing architectures.
• To analyze the integration of quantum computing with classical HPC and machine learning for efficient materials simulation.
• To address the limitations of current NISQ devices, including qubit count and noise, in simulating quantum materials.
• To propose a framework for optimizing both discrete and continuous variables in metamaterial design using quantum algorithms.
• To establish criteria for identifying high-impact use cases in materials science where quantum advantage is feasible.

Proposed method

• The paper employs a multi-level analysis spanning fundamental quantum algorithms, hybrid quantum-classical architecture design, and classical simulation of quantum systems for verification.
• It evaluates classical workloads associated with quantum algorithms, particularly focusing on the computational bottlenecks in simulating many-body quantum systems.
• The authors propose a hybrid quantum-classical algorithm capable of optimizing both discrete and continuous design parameters in metamaterials.
• They leverage classical HPC to simulate quantum systems for approximate verification and to identify computationally hard cases.
• The framework integrates machine learning to accelerate the discovery of optimal metamaterial structures.
• The approach emphasizes co-design of quantum and classical workloads to maximize performance in quantum-centric supercomputing centers.

Experimental results

Research questions

• RQ1 How can quantum computing be effectively integrated with classical HPC to achieve quantum advantage in materials science?
• RQ2 What are the key architectural and algorithmic challenges in scaling quantum simulations for materials on NISQ and fault-tolerant devices?
• RQ3 How can quantum algorithms be adapted to optimize continuous variables in materials design, given that standard quantum algorithms return binary outcomes?
• RQ4 Which materials science use cases are most promising for demonstrating quantum advantage in quantum-centric supercomputing?
• RQ5 What criteria can be used to identify and prioritize high-impact applications in materials science for quantum computing?

Key findings

• The integration of quantum computing, HPC, and machine learning presents a viable pathway to accelerate materials discovery, particularly for complex metamaterials.
• Current NISQ devices are limited by qubit count and noise, necessitating hybrid quantum-classical workloads and classical simulation for verification.
• A novel quantum algorithm is proposed that enables simultaneous optimization of discrete and continuous design variables, overcoming a key limitation of standard quantum algorithms.
• Classical simulations of quantum systems are essential for identifying hard use cases and validating quantum algorithm performance.
• Quantum-centric supercomputing centers that co-design quantum and classical workloads are critical for achieving scalable and practical quantum advantage in materials science.
• The paper establishes a framework for selecting high-impact materials science applications where quantum advantage is most likely to emerge.