Fitting Autoregressive Graph Generative Models through Maximum Likelihood Estimation
Fitting Autoregressive Graph Generative Models through Maximum Likelihood Estimation
复制标题
DOI：
--
复制
发表时间：
2023
期刊：
影响因子：
--
通讯作者：
Xuhong Han;Xiaohui Chen;Francisco J. R. Ruiz;Liping Liu
中科院分区：
文献类型：
--
作者：
Xuhong Han;Xiaohui Chen;Francisco J. R. Ruiz;Liping Liu
我们考虑的问题，通过最大似然估计（MLE）拟合自回归图生成模型。MLE对于图自回归模型是难以处理的，因为图中的节点可以任意重新排序;因此，确切的可能性涉及导致相同图的所有可能节点顺序的总和。在这项工作中，我们通过最大化变分界限来拟合图模型，该变分界限是通过首先导出图上的联合概率和自回归过程的节点顺序来建立的。这种方法避免了指定ad-hoc节点顺序的需要，因为推理网络学习生成给定图的最可能节点序列。我们通过开发一个基于注意机制的图生成模型和一个基于路由搜索的推理网络来改进该方法。我们实证证明，通过变分推理拟合自回归图模型可以提高其定性和定量性能，改进的模型和推理网络进一步提高了性能。拟议模型的实施可在https://github.com/tufts-ml/Graph-Generation-MLE上公开获得。
We consider the problem of fitting autoregressive graph generative models via maximum likelihood estimation (MLE). MLE is intractable for graph autoregressive models because the nodes in a graph can be arbitrarily reordered; thus the exact likelihood involves a sum over all possible node orders leading to the same graph. In this work, we fit the graph models by maximizing a variational bound, which is built by first deriving the joint probability over the graph and the node order of the autoregressive process. This approach avoids the need to specify ad-hoc node orders, since an inference network learns the most likely node sequences that have generated a given graph. We improve the approach by developing a graph generative model based on attention mechanisms and an inference network based on routing search. We demonstrate empirically that fitting autoregressive graph models via variational inference improves their qualitative and quantitative performance, and the improved model and inference network further boost the performance. The implementation of the proposed model is publicly available at https://github.com/tufts-ml/Graph-Generation-MLE.