An Unpooling Layer for Graph Generation
An Unpooling Layer for Graph Generation
复制标题
DOI：
10.48550/arxiv.2206.01874
复制
发表时间：
2022-06
期刊：
影响因子：
--
通讯作者：
Yi Guo;Dongmian Zou;Gilad Lerman
中科院分区：
文献类型：
--
作者：
Yi Guo;Dongmian Zou;Gilad Lerman
我们提出了一种新颖的、可训练的图分解层来有效地生成图。给定一个带有特征的图形，解合层将该图形放大并学习其所需的新结构和特征。由于解合层是可训练的，因此它既可以用于变分自动编码器的解码器中，也可以用于生成性对抗网络(GAN)的生成器中。我们证明了非池化图保持连通，且任何连通图都可以从3结点图中顺序地去池化。我们在GaN产生器内施加去池层。由于研究最多的图生成实例是分子生成，因此我们在此背景下测试我们的想法。使用QM9和ZINE数据集，我们演示了通过使用解池层而不是基于邻接矩阵的方法所获得的改进。
We propose a novel and trainable graph unpooling layer for effective graph generation. Given a graph with features, the unpooling layer enlarges this graph and learns its desired new structure and features. Since this unpooling layer is trainable, it can be applied to graph generation either in the decoder of a variational autoencoder or in the generator of a generative adversarial network (GAN). We prove that the unpooled graph remains connected and any connected graph can be sequentially unpooled from a 3-nodes graph. We apply the unpooling layer within the GAN generator. Since the most studied instance of graph generation is molecular generation, we test our ideas in this context. Using the QM9 and ZINC datasets, we demonstrate the improvement obtained by using the unpooling layer instead of an adjacency-matrix-based approach.