D-VAE: A Variational Autoencoder for Directed Acyclic Graphs
D-VAE: A Variational Autoencoder for Directed Acyclic Graphs
复制标题
DOI：
--
复制
发表时间：
2019-04
期刊：
影响因子：
--
通讯作者：
Muhan Zhang;Shali Jiang;Zhicheng Cui;R. Garnett;Yixin Chen
中科院分区：
文献类型：
--
作者：
Muhan Zhang;Shali Jiang;Zhicheng Cui;R. Garnett;Yixin Chen
图结构化数据在真实的世界中非常丰富。在不同的图类型中，有向无环图（DAG）是机器学习研究人员特别感兴趣的，因为许多机器学习模型都是在DAG上实现的，包括神经网络和贝叶斯网络。在本文中，我们研究了DAG的深度生成模型，并提出了一种新的DAG变分自动编码器（D-VAE）。为了将DAG编码到潜在空间中，我们利用图神经网络。我们提出了一个异步的消息传递计划，允许编码的DAG上的计算，而不是使用现有的同步消息传递计划来编码本地图结构。我们证明了我们提出的DVAE的有效性，通过两个任务：神经结构搜索和贝叶斯网络结构学习。实验表明，我们的模型不仅产生新颖和有效的DAG，但也产生了一个光滑的潜在空间，有利于搜索DAG具有更好的性能，通过贝叶斯优化。
Graph structured data are abundant in the real world. Among different graph types, directed acyclic graphs (DAGs) are of particular interest to machine learning researchers, as many machine learning models are realized as computations on DAGs, including neural networks and Bayesian networks. In this paper, we study deep generative models for DAGs, and propose a novel DAG variational autoencoder (D-VAE). To encode DAGs into the latent space, we leverage graph neural networks. We propose an asynchronous message passing scheme that allows encoding the computations on DAGs, rather than using existing simultaneous message passing schemes to encode local graph structures. We demonstrate the effectiveness of our proposed DVAE through two tasks: neural architecture search and Bayesian network structure learning. Experiments show that our model not only generates novel and valid DAGs, but also produces a smooth latent space that facilitates searching for DAGs with better performance through Bayesian optimization.