BriefGPT.xyz
May, 2019
路径增强图变换器网络
Path-Augmented Graph Transformer Network
HTML
PDF
Benson Chen, Regina Barzilay, Tommi Jaakkola
TL;DR
提出了一种基于路径增强的图转换网络(PAGTN)，用于学习分子表示，并将其与基于图卷积网络(GCN)的模型进行比较，在分子属性预测方面表现更好，包括量子化学(QM7，QM8，QM9)，物理化学(ESOL，亲脂)和生物化学(BACE，BBBP)数据集。
Abstract
Much of the recent work on learning
molecular representations
has been based on
graph convolution networks
(GCN). These models rely on local aggregation operations and can therefore miss higher-order graph proper
→