Variational Autoencoders for Highly Multivariate Spatial Point Processes Intensities
Variational Autoencoders for Highly Multivariate Spatial Point Processes Intensities
复制标题
DOI：
--
复制
发表时间：
2020-04
期刊：
影响因子：
--
通讯作者：
Baichuan Yuan;Xiaowei Wang;Jianxin Ma;Chang Zhou;A. Bertozzi;Hongxia Yang
中科院分区：
文献类型：
--
作者：
Baichuan Yuan;Xiaowei Wang;Jianxin Ma;Chang Zhou;A. Bertozzi;Hongxia Yang
多元空间点过程模型可以描述空间上的异位数据。然而，由于维数的诅咒，高多元强度在计算上具有挑战性。为了弥补这一差距，我们引入了一个基于聚类的隐变量模型，该模型通过变分自编码器（VAE）实现了一个有效的推理过程。我们还证明了该模型是基于vae的协同过滤模型的推广。这导致了空间点过程模型在推荐系统中的有趣应用。实验结果表明，该方法在合成数据和实际数据集上都是有效的。
Multivariate spatial point process models can describe heterotopic data over space. However, highly multivariate intensities are computationally challenging due to the curse of dimensionality. To bridge this gap, we introduce a declustering based hidden variable model that leads to an efficient inference procedure via a variational autoencoder (VAE). We also prove that this model is a generalization of the VAE-based model for collaborative filtering. This leads to an interesting application of spatial point process models to recommender systems. Experimental results show the method's utility on both synthetic data and real-world data sets.