Abstract Integrating AI into materials science accelerates material discovery. This study develops an Ising model for composite material optimization using machine learning and solves it efficiently with an annealing computer. Unlike traditional methods limited to directly representable problems, our approach converts machine learning into Ising models. It includes a transformation unit to binarize variables, a training unit to learn an Ising model, and an output unit to optimize using the trained model. With an annealing computer, we achieve speeds tens of thousands of times faster than exhaustive methods. Compositional constraints are incorporated as penalty functions to ensure valid combinations. This novel method transforms machine learning predictions into Ising models, enabling rapid optimization and advancing AI applications in materials science. Graphical abstract
Abstract Integrating AI into materials science accelerates material discovery. This study develops an Ising model for composite material optimization using machine learning and solves it efficiently with an annealing computer. Unlike traditional methods limited to directly representable problems, our approach converts machine learning into Ising models. It includes a transformation unit to binarize variables, a training unit to learn an Ising model, and an output unit to optimize using the trained model. With an annealing computer, we achieve speeds tens of thousands of times faster than exhaustive methods. Compositional constraints are incorporated as penalty functions to ensure valid combinations. This novel method transforms machine learning predictions into Ising models, enabling rapid optimization and advancing AI applications in materials science. Graphical abstract
Artificial Intelligence is rapidly transforming materials science and engineering, offering powerful tools to navigate complexity, accelerate discovery, and optimize material design in ways previously unattainable. Driven by the accelerating pace of algorithmic advancements and increasing data availability, AI is becoming an essential competency for materials researchers. This review provides a comprehensive and structured overview of the current landscape, synthesizing recent advancements and methodologies for materials scientists seeking to effectively leverage these data-driven techniques. We survey the spectrum of machine learning approaches, from traditional algorithms to advanced deep learning architectures, including CNNs, GNNs, and Transformers, alongside emerging generative AI and probabilistic models such as Gaussian Processes for uncertainty quantification. The review also examines the pivotal role of data in this field, emphasizing how effective representation and featurization strategies, spanning compositional, structural, image-based, and language-inspired approaches, combined with appropriate preprocessing, fundamentally underpin the performance of machine learning models in materials research. Persistent challenges related to data quality, quantity, and standardization, which critically impact model development and application in materials science and engineering, are also addressed.
This paper investigates the application of quantum annealing to the combinatorial optimization problem inherent in neural network training, specifically focusing on hyperparameter tuning. Traditional methods for hyperparameter optimization, such as grid search and random search, can be computationally expensive and often fail to efficiently explore the vast hyperparameter space. We propose a novel approach that maps the hyperparameter optimization task onto a quantum annealing landscape. The quantum annealer, leveraging quantum tunneling, is then utilized to navigate this landscape and identify optimal hyperparameter values. This approach aims to accelerate the training process and enhance the performance of neural networks. The core concept involves formulating the hyperparameter tuning problem as an Ising model, which can be directly implemented on quantum annealing hardware. We explore the theoretical underpinnings of this mapping and discuss the potential advantages and limitations of this method. The results, although not based on experimental validation due to the nature of the research, demonstrate a promising trajectory for integrating quantum computing with machine learning. Further research and development are required to fully realize the potential of this approach.
Materials science has historically depended on a blend of experimental methods and theoretical modelling to identify and create new materials with specific properties. Nevertheless, these processes may require considerable time and resources, and are frequently constrained by the intricacy of material systems. The rise of artificial intelligence (AI), particularly machine learning, has revolutionised materials science by offering powerful tools that accelerate the discovery, design, and characterisation of novel materials. This chapter emphasises the latest developments in AI applications in materials science for drug discovery. Artificial Intelligence is proficient at analysing intricate data, enhancing processes, and developing drug candidates, whereas quantum systems enable unparalleled molecular simulations, highly sensitive sensing, and accurate physical control. Applications in drug discovery are emphasised, encompassing molecular property prediction and molecular generation. This chapter focuses on technologies such as Nanomaterials, Biomaterials, Polymers, Metal-Organic Frameworks (MOFs), Hydrogels, and Smart Materials. This chapter emphasises the advantages of quantum technology in drug discovery: enhanced accuracy in molecular simulations, Accelerated drug screening, better comprehension of reactions, and Tailored medicine. Additionally, the challenges include: hardware limitations, the high cost of error correction, the maturity of algorithms, integration with classical methods, and issues related to cost and accessibility.
Data scarcity remains a central challenge in materials discovery, where finding meaningful descriptors and tuning models for generalization is critical but inherently a discrete optimization problem prone to multiple local minima confounding the true optimal state. Classical methods often get trapped in these minima, while quantum annealing can escape them via quantum fluctuations, including tunneling, that overcome narrow energy barriers. We present a quantum-assisted machine-learning (QaML) framework that employs quantum annealing to address these combinatorial optimization challenges through feature selection, support-vector training formulated in QUBO form for classification and regression, and a new QUBO-based neural-network pruning formulation. Recursive batching enables quantum annealing to handle large feature spaces beyond current qubit limits, while quantum-pruned networks exhibit superior generalization over classical methods, suggesting that quantum annealing preferentially samples flatter, more stable regions of the loss landscape. Applied to high-entropy alloys (HEAs), a data-limited but compositionally complex testbed, the framework integrates models for fracture-strain classification and yield-strength regression under physics-based constraints. The framework identified and experimentally validated Al8Cr38Fe50Mn2Ti2 (at.%), a single-phase BCC alloy exhibiting a 0.2 % yield strength of 568 MPa, greater than 40 % compressive strain without fracture, and a critical current density in reducing acid nearly an order of magnitude lower than 304 stainless steel. These results establish QA as a practical route to overcome classical optimization limits and accelerate materials discovery.
Quantum machine learning (QML) is transforming the field of machine learning (ML), which typically utilizes a quantum computing such as gate model-based and annealing-based computing. Recently, Ising solver chips have been developed to emulate the superposition and entanglement properties of quantum physics using the current silicon manufacturing technologies. The chips have potential to solve complex combinatorial optimization problems (COP) with significantly low power consumption and price. This paper proposes quantum-inspired ML (QiML) framework that provides an approach to train the traditional ML models using a physics-based Ising solver chip to develop QiMLs. To validate the feasibility of the proposed framework, a quantum-inspired support vector machine (QiSVM) is developed and the results are compared with a quantum SVM trained by a commercial D-Wave quantum annealer and a conventional SVM to solve real-world problems. The proposed QiML framework will open new frontiers for advancing ML technologies for industry cyber-physical system applications.
The design of next-gen materials has undergone remarkable progress in recent years, as evidenced by the emergence of automated platforms combining artificial intelligence (AI)-driven synthesis planning and robotics for execution. In this Mini-Review, we analyze how synergistic approaches that combine driven quantum dynamics, AI/machine learning, and quantum computing accelerate the discovery and design process of quantum materials with enhanced properties and novel functionalities. Building on the capabilities of each of the three methods, synergistic approaches can provide access to the materials' response to time-dependent fields, enable the rapid exploration of vast design spaces, and identify novel quantum phases and materials with optimal properties. We examine recent successes in next-gen materials science for quantum batteries, colloidal quantum dots solar cells, quantum phototransistors, rare-earth-free materials, and applications in quantum information processing. We conclude with a discussion of recent research efforts in AI-for-quantum computing and quantum machine learning for next-gen materials discovery.