气候变化领域集成服务门户(CCPortal): PhAST: Physics-Aware, Scalable, and Task-Specific GNNs for Accelerated Catalyst Design

CCPortal

	PhAST: Physics-Aware, Scalable, and Task-Specific GNNs for Accelerated Catalyst Design
	Duval, Alexandre; Schmidt, Victor; Miret, Santiago; Bengio, Yoshua; Hernandez-Garcia, Alex; Rolnick, David
发表日期	2024
ISSN	1532-4435
起始页码	25
卷号	25
英文摘要	Mitigating the climate crisis requires a rapid transition towards lower -carbon energy. Catalyst materials play a crucial role in the electrochemical reactions involved in numerous industrial processes key to this transition, such as renewable energy storage and electrofuel synthesis. To reduce the energy spent on such activities, we must quickly discover more efficient catalysts to drive electrochemical reactions. Machine learning (ML) holds the potential to efficiently model materials properties from large amounts of data, accelerating electrocatalyst design. The Open Catalyst Project OC20 dataset was constructed to that end. However, ML models trained on OC20 are still neither scalable nor accurate enough for practical applications. In this paper, we propose task -specific innovations applicable to most architectures, enhancing both computational efficiency and accuracy. This includes improvements in (1) the graph creation step, (2) atom representations, (3) the energy prediction head, and (4) the force prediction head. We describe these contributions, referred to as PhAST, and evaluate them thoroughly on multiple architectures. Overall, PhAST improves energy MAE by 4 to 42% while dividing compute time by 3 to 8x depending on the targeted task/model. PhAST also enables CPU training, leading to 40x speedups in highly parallelized settings. Python package: https://phast.readthedocs.io.
英文关键词	climate change; scientific discovery; material modeling; graph neural networks; electrocatalysts
语种	英语
WOS研究方向	Automation & Control Systems ; Computer Science
WOS类目	Automation & Control Systems ; Computer Science, Artificial Intelligence
WOS记录号	WOS:001203728000001
来源期刊	JOURNAL OF MACHINE LEARNING RESEARCH
文献类型	期刊论文
条目标识符	http://gcip.llas.ac.cn/handle/2XKMVOVA/294748
作者单位	Inria; Universite Paris Saclay; Universite de Montreal; Intel Corporation; McGill University
推荐引用方式 GB/T 7714	Duval, Alexandre,Schmidt, Victor,Miret, Santiago,et al. PhAST: Physics-Aware, Scalable, and Task-Specific GNNs for Accelerated Catalyst Design[J],2024,25.
APA	Duval, Alexandre,Schmidt, Victor,Miret, Santiago,Bengio, Yoshua,Hernandez-Garcia, Alex,&Rolnick, David.(2024).PhAST: Physics-Aware, Scalable, and Task-Specific GNNs for Accelerated Catalyst Design.JOURNAL OF MACHINE LEARNING RESEARCH,25.
MLA	Duval, Alexandre,et al."PhAST: Physics-Aware, Scalable, and Task-Specific GNNs for Accelerated Catalyst Design".JOURNAL OF MACHINE LEARNING RESEARCH 25(2024).