论文标题
图形各向异性扩散
Graph Anisotropic Diffusion
论文作者
论文摘要
传统的图形神经网络（GNN）依赖消息传递，这等于邻居特征的置换局部局部聚合。这样的过程是各向同性的，图表上没有“方向”的概念。我们提出了一种新的GNN体系结构，称为图形各向异性扩散。我们的模型在线性扩散之间进行交替，为其提供了封闭形式的解决方案，以及局部各向异性过滤器，以获得有效的多跳各向异性核。我们在两个常见的分子属性预测基准（锌和QM9）上测试模型，并显示其竞争性能。
Traditional Graph Neural Networks (GNNs) rely on message passing, which amounts to permutation-invariant local aggregation of neighbour features. Such a process is isotropic and there is no notion of `direction' on the graph. We present a new GNN architecture called Graph Anisotropic Diffusion. Our model alternates between linear diffusion, for which a closed-form solution is available, and local anisotropic filters to obtain efficient multi-hop anisotropic kernels. We test our model on two common molecular property prediction benchmarks (ZINC and QM9) and show its competitive performance.
