About this Research Topic
Glasses are often undervalued and considered economical, low-tech materials for windows and beverage containers. However, while many commercial glasses are indeed cost-effective (and this is one of their industrial strengths), they are far from being low-tech. In recent decades, the development of new glassy materials has been instrumental in addressing critical global challenges in areas such as energy, medicine, and advanced communication systems. The versatility of glasses stems from their amorphous nature, which allows for compositions that do not need to satisfy stoichiometric requirements, as in the case of crystals. This attribute enables flexibility in discovering glasses with unique combinations of properties that are not attainable with other materials.
The lack of long-range order in glasses has impeded a complete understanding of the relationship between composition, structure, and properties. Traditional methods of designing commercial glasses involve a time-consuming ‘trial-and-error’ approach. However, with the growing demand for glasses with specific properties for various applications, there is a need for more efficient methods. Using machine learning (ML) or artificial intelligence (AI) techniques can help bridge this gap and enable a ‘material-by-design’ approach instead of the current ‘trial-and-error’ method.
By processing large amounts of data, both experimental and simulated, ML or AI can quickly establish relationships and predict the properties or compositions of new glasses in a more economical and environment-friendly manner.
This Research Topic aims to collect and examine the latest advancements and applications of ML techniques in understanding the relationship between composition and properties from both experimental and computational perspectives.
Original research and review articles focusing on the application of machine learning techniques to glasses and glass-ceramics are welcome.
Of specific interest are studies that provide novel and insightful information about material behaviour, properties, and phenomena using ML, new ML techniques never applied to glass science, as well as their application to the discover of new glass compositions.
Suitable topics include but are not limited to:
- Multicomponent oxide, chalcogenide, and metallic glasses.
- Application of ML algorithm toward improved glass composition design, property prediction, and the discovering the relationship between composition and atomic structure.
- Machine learning potentials (or force fields) for molecular dynamics simulations of glasses.
- Comparison of ML algorithms in different fields related to glassy materials.
- Current state-of-the-art and future development in the application of ML techniques to glass and glass-ceramics.
- Combining high-throughput simulations with ML for structure¬–property relationships of glasses.
- Self-driving laboratory for high-throughput glass synthesis and testing.
- Natural language processing for information extraction on glass compositions and properties from scientific literature.
- Computer vision for accelerated inference of glass properties from experimental data.
- Advanced ML algorithms such as reinforcement learning and transfer learning for discovering the fundamentals of glassy states and accelerated glass discovery.
The lack of long-range order in glasses has impeded a complete understanding of the relationship between composition, structure, and properties. Traditional methods of designing commercial glasses involve a time-consuming ‘trial-and-error’ approach. However, with the growing demand for glasses with specific properties for various applications, there is a need for more efficient methods. Using machine learning (ML) or artificial intelligence (AI) techniques can help bridge this gap and enable a ‘material-by-design’ approach instead of the current ‘trial-and-error’ method.
By processing large amounts of data, both experimental and simulated, ML or AI can quickly establish relationships and predict the properties or compositions of new glasses in a more economical and environment-friendly manner.
This Research Topic aims to collect and examine the latest advancements and applications of ML techniques in understanding the relationship between composition and properties from both experimental and computational perspectives.
Original research and review articles focusing on the application of machine learning techniques to glasses and glass-ceramics are welcome.
Of specific interest are studies that provide novel and insightful information about material behaviour, properties, and phenomena using ML, new ML techniques never applied to glass science, as well as their application to the discover of new glass compositions.
Suitable topics include but are not limited to:
- Multicomponent oxide, chalcogenide, and metallic glasses.
- Application of ML algorithm toward improved glass composition design, property prediction, and the discovering the relationship between composition and atomic structure.
- Machine learning potentials (or force fields) for molecular dynamics simulations of glasses.
- Comparison of ML algorithms in different fields related to glassy materials.
- Current state-of-the-art and future development in the application of ML techniques to glass and glass-ceramics.
- Combining high-throughput simulations with ML for structure¬–property relationships of glasses.
- Self-driving laboratory for high-throughput glass synthesis and testing.
- Natural language processing for information extraction on glass compositions and properties from scientific literature.
- Computer vision for accelerated inference of glass properties from experimental data.
- Advanced ML algorithms such as reinforcement learning and transfer learning for discovering the fundamentals of glassy states and accelerated glass discovery.
Keywords: Glass, Glass-Ceramic, Machine Learning, Artificial Intelligence, Material-by-design, Properties prediction
Important Note: All contributions to this Research Topic must be within the scope of the section and journal to which they are submitted, as defined in their mission statements. Frontiers reserves the right to guide an out-of-scope manuscript to a more suitable section or journal at any stage of peer review.
