arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2102.03150cs.LGphysics.chem-ph
收藏

等变信息传递用于张量性质和分子光谱的预测

Equivariant message passing for the prediction of tensorial properties and molecular spectra

Technische Universität Berlin(柏林工业大学);Berlin Institute for the Foundations of Learning and Data(柏林学习与数据基础研究所)

Kristof T. Schütt, Oliver T. Unke, Michael Gastegger

摘要:

消息传递神经网络已成为一种选择的方法,用于学习图,特别是化学性质的预测和分子动力学研究的加速。虽然它们很容易扩展到大的训练数据集,但以前的方法已被证明比内核方法的数据效率低。我们确定的一个主要原因不变表示的限制,并扩展到旋转等变表示的消息传递制定。在此基础上,我们提出了极化原子相互作用神经网络(PaiNN),并对常用的分子基准进行了改进,同时减小了模型规模和推理时间。我们利用PaiNN获得的等变atomwise表示来预测张量性质。最后,我们将其应用于分子光谱的模拟,与电子结构参考相比,实现了4-5个数量级的加速。

英文摘要:

Message passing neural networks have become a method of choice for learning on graphs, in particular the prediction of chemical properties and the acceleration of molecular dynamics studies. While they readily scale to large training data sets, previous approaches have proven to be less data efficient than kernel methods. We identify limitations of invariant representations as a major reason and extend the message passing formulation to rotationally equivariant representations. On this basis, we propose the polarizable atom interaction neural network (PaiNN) and improve on common molecule benchmarks over previous networks, while reducing model size and inference time. We leverage the equivariant atomwise representations obtained by PaiNN for the prediction of tensorial properties. Finally, we apply this to the simulation of molecular spectra, achieving speedups of 4-5 orders of magnitude compared to the electronic structure reference.

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。

↑