Graph-based deep learning frameworks for molecules and solid-state materials

Gong, Weiyi; Yan, Qimin

doi:10.1016/j.commatsci.2021.110332

Cited by 16 publications

(10 citation statements)

References 37 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This symmetry makes it necessary to accommodate periodic boundary conditions because of which conventional models that are used for these materials cannot directly be applied to molecules or atomic environments. 67 From a machine-learning perspective, the difference between these materials lies in the relevant information they present that needs to be extracted through a process known as “feature extraction” or “feature engineering”. These features are then fed into machine-learning models to make inferences and/or predictions.…”

Section: Applications Of Generative Models In Materials Sciencementioning

confidence: 99%

“…Here, crystals are represented graphically through their unit cells, where the atoms that make up the unit cell act as nodes and the distances between these atoms (nodes) serve as edges of the cell. 67 Newer techniques account for periodic boundary conditions, as is the case with SchNet, a deep-learning architecture based on deep tensor neural networks that model complex atomic interactions while incorporating periodic boundary conditions, in order to predict the formation energies of bulk crytals. 68 Detailed discussions on feature engineering and advanced techniques to carry it out, especially for solid-state crystalline materials, have been provided by Schmidt et al 23 and Gong et al 67 …”

Section: Applications Of Generative Models In Materials Sciencementioning

confidence: 99%

“…Fundamentally, solid-state crystal materials are composed of periodic arrays of repeating units, leading to symmetries at both a local and global level. This symmetry makes it necessary to accommodate periodic boundary conditions because of which conventional models that are used for these materials cannot directly be applied to molecules or atomic environments . From a machine-learning perspective, the difference between these materials lies in the relevant information they present that needs to be extracted through a process known as “feature extraction” or “feature engineering”.…”

Section: Applications Of Generative Models In Materials Sciencementioning

confidence: 99%

“…An emerging technique for the feature engineering of solid-state crystalline materials is through the use of graph-based networks. Here, crystals are represented graphically through their unit cells, where the atoms that make up the unit cell act as nodes and the distances between these atoms (nodes) serve as edges of the cell . Newer techniques account for periodic boundary conditions, as is the case with SchNet, a deep-learning architecture based on deep tensor neural networks that model complex atomic interactions while incorporating periodic boundary conditions, in order to predict the formation energies of bulk crytals .…”

Section: Applications Of Generative Models In Materials Sciencementioning

confidence: 99%

“…67 Newer techniques account for periodic boundary conditions, as is the case with SchNet, a deeplearning architecture based on deep tensor neural networks that model complex atomic interactions while incorporating periodic boundary conditions, in order to predict the formation energies of bulk crytals. 68 Detailed discussions on feature engineering and advanced techniques to carry it out, especially for solid-state crystalline materials, have been provided by Schmidt et al 23 and Gong et al 67 3.1. Biomaterials and Organic Drug-like Molecules.…”

Section: Applications Of Generative Models In Materials Sciencementioning

confidence: 99%

See 4 more Smart Citations

A Generative Approach to Materials Discovery, Design, and Optimization

Menon

Ranganathan

2022

ACS Omega

View full text Add to dashboard Cite

Despite its potential to transform society, materials research suffers from a major drawback: its long research timeline. Recently, machine-learning techniques have emerged as a viable solution to this drawback and have shown accuracies comparable to other computational techniques like density functional theory (DFT) at a fraction of the computational time. One particular class of machine-learning models, known as “generative models”, is of particular interest owing to its ability to approximate high-dimensional probability distribution functions, which in turn can be used to generate novel data such as molecular structures by sampling these approximated probability distribution functions. This review article aims to provide an in-depth understanding of the underlying mathematical principles of popular generative models such as recurrent neural networks, variational autoencoders, and generative adversarial networks and discuss their state-of-the-art applications in the domains of biomaterials and organic drug-like materials, energy materials, and structural materials. Here, we discuss a broad range of applications of these models spanning from the discovery of drugs that treat cancer to finding the first room-temperature superconductor and from the discovery and optimization of battery and photovoltaic materials to the optimization of high-entropy alloys. We conclude by presenting a brief outlook of the major challenges that lie ahead for the mainstream usage of these models for materials research.

show abstract

Section: Applications Of Generative Models In Materials Sciencementioning

confidence: 99%

Section: Applications Of Generative Models In Materials Sciencementioning

confidence: 99%

Section: Applications Of Generative Models In Materials Sciencementioning

confidence: 99%

Section: Applications Of Generative Models In Materials Sciencementioning

confidence: 99%

Section: Applications Of Generative Models In Materials Sciencementioning

confidence: 99%

See 3 more Smart Citations

A Generative Approach to Materials Discovery, Design, and Optimization

Menon

Ranganathan

2022

ACS Omega

View full text Add to dashboard Cite

show abstract

GenAI for Scientific Discovery in Electrochemical Energy Storage: State‐of‐the‐Art and Perspectives from Nano‐ and Micro‐Scale

Li,

You

2024

Small

View full text Add to dashboard Cite

The transition to electric vehicles (EVs) and the increased reliance on renewable energy sources necessitate significant advancements in electrochemical energy storage systems. Fuel cells, lithium‐ion batteries, and flow batteries play a key role in enhancing the efficiency and sustainability of energy usage in transportation and storage. Despite their potential, these technologies face limitations such as high costs, material scarcity, and efficiency challenges. This research introduces a novel integration of Generative AI (GenAI) within electrochemical energy storage systems to address these issues. By leveraging advanced GenAI techniques like Generative Adversarial Networks, autoencoders, diffusion and flow‐based models, and multimodal large language models, this paper demonstrates significant improvements in material discovery, battery design, performance prediction, and lifecycle management across different types of electrochemical storage systems. The research further emphasizes the importance of nano‐ and micro‐scale interactions, providing detailed insights into optimizing these interactions for improved efficiency and longevity. Additionally, the paper discusses the challenges and future directions for integrating GenAI in energy storage research, highlighting the importance of data quality, model transparency, workflow integration, scalability, and ethical considerations. By addressing these aspects, this research sets a new benchmark for the use of GenAI in battery development, promoting sustainable, efficient, and safer energy solutions.

show abstract