arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2406.04713cs.LGcond-mat.mtrl-scics.AIphysics.comp-phstat.ML
收藏

FlowMM:基于黎曼流匹配生成材料

FlowMM: Generating Materials with Riemannian Flow Matching

University of Amsterdam(阿姆斯特丹大学);FAIR, Meta AI(Meta AI FAIR)

Benjamin Kurt Miller, Ricky T. Q. Chen, Anuroop Sriram, Brandon M Wood

AI总结:

FlowMM通过将黎曼流匹配推广到晶体对称性场景,提出一对生成模型,在晶体结构预测和新材料发现任务上以约3倍效率达到最先进性能。

摘要:

晶体材料是下一代技术中的基础组成部分,然而对其分布进行建模带来了独特的计算挑战。在周期性晶格中所有可能的原子排布方式里,只有极小的比例是热力学稳定的,而热力学稳定性是判断材料能否在实验中实现的关键指标。该领域的两个基本任务是:(a)预测已知元素组成的稳定晶体结构,以及(b)提出新的元素组成及其稳定结构。我们提出了FlowMM,这是一对生成模型,在这两项任务上均取得了最先进的性能,同时比现有竞争方法更高效、更灵活。我们将黎曼流匹配(Riemannian Flow Matching)推广到适合晶体固有对称性的情形:平移、旋转、置换和周期性边界条件。我们的框架允许自由选择流基分布,与扩散模型相比,极大地简化了学习晶体结构的问题。除了标准基准测试之外,我们还通过量子化学计算验证了FlowMM生成的结构,结果表明,在寻找稳定材料方面,就积分步数而言,FlowMM比之前的开放方法效率约高3倍。

英文摘要:

Crystalline materials are a fundamental component in next-generation technologies, yet modeling their distribution presents unique computational challenges. Of the plausible arrangements of atoms in a periodic lattice only a vanishingly small percentage are thermodynamically stable, which is a key indicator of the materials that can be experimentally realized. Two fundamental tasks in this area are to (a) predict the stable crystal structure of a known composition of elements and (b) propose novel compositions along with their stable structures. We present FlowMM, a pair of generative models that achieve state-of-the-art performance on both tasks while being more efficient and more flexible than competing methods. We generalize Riemannian Flow Matching to suit the symmetries inherent to crystals: translation, rotation, permutation, and periodic boundary conditions. Our framework enables the freedom to choose the flow base distributions, drastically simplifying the problem of learning crystal structures compared with diffusion models. In addition to standard benchmarks, we validate FlowMM's generated structures with quantum chemistry calculations, demonstrating that it is about 3x more efficient, in terms of integration steps, at finding stable materials compared to previous open methods.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑