论文标题
无监督的图神经网络揭示了无序系统中的结构 - 动力学相关性
Unsupervised Graph Neural Network Reveals the Structure--Dynamics Correlation in Disordered Systems
论文作者
论文摘要
学习结构 - 无序系统中的动力学相关性是一个长期存在的问题。在这里，我们使用图形神经网络（GNN）使用无监督的机器学习来研究无序系统中的局部结构。我们在2D二进制A65B35 LJ玻璃上测试了我们的方法，并提取与液体，过冷和玻璃状态相对应的以不同冷却速率的结构。 GNN以无监督的方式学到的原子的邻居表示，当被聚集时，揭示了具有不同势能的局部结构。这些簇在结构中表现出动态异质性，并与其局部能量景观一致。总的来说，本研究表明，无监督的图嵌入可以揭示结构的结构相关性。
Learning the structure--dynamics correlation in disordered systems is a long-standing problem. Here, we use unsupervised machine learning employing graph neural networks (GNN) to investigate the local structures in disordered systems. We test our approach on 2D binary A65B35 LJ glasses and extract structures corresponding to liquid, supercooled and glassy states at different cooling rates. The neighborhood representation of atoms learned by a GNN in an unsupervised fashion, when clustered, reveal local structures with varying potential energies. These clusters exhibit dynamical heterogeneity in the structure in congruence with their local energy landscape. Altogether, the present study shows that unsupervised graph embedding can reveal the structure--dynamics correlation in disordered structures.
