深度学习导致了人工智能的范式转变,包括网络、文本和图像搜索、语音识别以及生物信息学,对化学物理的影响越来越大。机器学习特别是深度学习非常适合表示量子力学相互作用,能够对非线性势能面进行建模或增强对化合物空间的探索。在这里,我们介绍了深度学习架构SchNet,该架构专门设计用于通过使用连续滤波器卷积层来建模原子系统。我们通过准确预测\emph{molecules and materials}的化学空间中的一系列属性来展示SchNet的能力,其中我们的模型学习了元素周期表中原子类型的化学合理嵌入。最后,我们使用SchNet来预测小分子动力学模拟的势能面和节能力场,并对C $_{20}$ -富勒烯的量子力学性质进行了示例性研究,这在常规从头算分子动力学中是不可行的。
英文摘要:
Deep learning has led to a paradigm shift in artificial intelligence, including web, text and image search, speech recognition, as well as bioinformatics, with growing impact in chemical physics. Machine learning in general and deep learning in particular is ideally suited for representing quantum-mechanical interactions, enabling to model nonlinear potential-energy surfaces or enhancing the exploration of chemical compound space. Here we present the deep learning architecture SchNet that is specifically designed to model atomistic systems by making use of continuous-filter convolutional layers. We demonstrate the capabilities of SchNet by accurately predicting a range of properties across chemical space for \emph{molecules and materials} where our model learns chemically plausible embeddings of atom types across the periodic table. Finally, we employ SchNet to predict potential-energy surfaces and energy-conserving force fields for molecular dynamics simulations of small molecules and perform an exemplary study of the quantum-mechanical properties of C$_{20}$-fullerene that would have been infeasible with regular ab initio molecular dynamics.
Benchmarking machine-learned interatomic potentials for molecular infrared spectroscopy
Technical University of Munich(慕尼黑工业大学);Aalto University(阿尔托大学);Munich Center for Machine Learning (MCML)(慕尼黑机器学习中心);Munich Data Science Institute(慕尼黑数据科学研究所);Department of Chemistry and Material Science, Aalto University(阿尔托大学化学与材料科学系)
Benchmarking Machine-Learning Interatomic Potentials for Dynamical Stability in Inorganic Semiconductor Nanocrystals: A CdSe Case Study
BCMaterials, Basque Center for Materials, Applications, and Nanostructures, UPV/EHU(巴斯克材料、应用与纳米结构中心(BCMaterials));University of the Basque Country (UPV/EHU)(巴斯克大学);Ikerbasque Basque Foundation for Science(伊克拉巴斯克科学基金会)
Developing Machine-Learned Potentials for Coarse-Grained Molecular Simulations: Challenges and Pitfalls
Institute of Informatics and Telecommunications(信息学与电信研究所);Institute of Nanoscience and Nanotechnology(纳米科学与纳米技术研究所);National Centre for Scientific Research “Demokritos”(“德谟克利特”国家科学研究中心);SciFY P.N.P.C.(SciFY公共非营利公司);School of Chemical Engineering(化学工程学院);National Technical University of Athens(雅典国立技术大学)
Machine Learning for Screening Large Organic Molecules
有机半导体是廉价、可扩展和可持续的电子器件、发光二极管和光电子器件的有前途的材料。对于有机光电池来说,在广阔的化合物空间中寻找具有合适性质的化合物是一个挑战。例如,电离能应该适合太阳光的光谱,并且能级必须允许有效的电荷传输。在这里,开发了一种机器学习模型,用于快速准确地估计给定分子结构的HOMO和LUMO能量。它建立在SchNet模型(Schütt et al.(2018))的基础上,并增加了“Set 2Set”读出模块(Vinyals et al.(2016))。Set 2Set模块比求和和平均聚合具有更强的表达能力,更适合于所考虑的复杂量。大多数以前的模型都是在相当小的分子上训练和评估的。因此,第二个贡献是通过添加来自其他来源的更大分子并建立一致的训练/验证/测试分割来扩展机器学习方法的范围。作为第三个贡献,我们做了一个多任务分析,以解决来自不同理论层次的不同来源的问题。所有这三个贡献结合起来,使模型的准确性接近化学准确性。
Analyzing Atomic Interactions in Molecules as Learned by Neural Networks
BIFOLD - Berlin Institute for the Foundations of Learning and Data(柏林学习与数据基础研究所);Machine Learning Group, Berlin Institute of Technology(柏林工业大学机器学习组);Department of Physics and Materials Science, University of Luxembourg(卢森堡大学物理与材料科学系);Department of Mathematics and Computer Science, Free University of Berlin(柏林自由大学数学与计算机科学系);Google Deepmind, Berlin(谷歌DeepMind(柏林))
Generating equilibrium molecules with deep neural networks
Technische Universität Berlin(柏林工业大学)
发现具有理想性质的原子体系是化学和材料科学中的一个重大挑战。在这里,我们介绍了一种新的自回归卷积深度神经网络架构,它通过在三维空间中顺序放置原子来生成分子平衡结构。该模型用易于处理的条件概率估计了分子构型上的联合概率,条件概率仅取决于原子与其核电荷之间的距离。它将最先进的原子神经网络的概念与图像和语音的自回归生成模型相结合。我们证明,该架构是能够产生接近平衡的C $_7$ O $_2$ H $_{10}$的结构异构体的分子。
TERS-ABNet: A Deep Learning Approach for Automated Single-Molecule Structure Reconstruction with Atomic Precision from TERS Mapping
Hefei National Research Center for Physical Sciences at the Microscale and CAS Center for Excellence in Quantum Information and Quantum Physics, University of Science and Technology of China(中国科学技术大学合肥微尺度物质科学国家研究中心和量子信息科学研究院);Hefei National Laboratory, University of Science and Technology of China(中国科学技术大学合肥国家实验室);School of Physical Sciences and Department of Chemical Physics, University of Science and Technology of China(中国科学技术大学物理科学学院化学物理系)
Accurate Prediction of Free Solvation Energy of Organic Molecules via Graph Attention Network and Message Passing Neural Network from Pairwise Atomistic Interactions
University of Michigan(密歇根大学);Stanford University(斯坦福大学)