arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 1705.08051cs.LGstat.ML
收藏

深层生成点过程模型的Wasserstein学习

Wasserstein Learning of Deep Generative Point Process Models

Shanghai Jiao Tong University(上海交通大学);Georgia Institute of Technology(佐治亚理工学院);Georgia State University(佐治亚州立大学);IBM

Shuai Xiao, Mehrdad Farajtabar, Xiaojing Ye, Junchi Yan, Le Song, Hongyuan Zha

摘要:

点过程由于其良好的数学基础和对各种现实世界现象的建模能力,在异步序列数据建模中变得非常流行。目前,它们通常通过强度函数来表征,这限制了模型的表达能力,这是由于在实践中使用的参数形式上的不切实际的假设。此外,它们是通过最大似然方法,这是容易失败的多模态分布的序列。在本文中,我们提出了一个强度自由的方法点过程建模,将滋扰过程的目标之一。此外,我们训练模型使用点过程之间的似然无杠杆Wasserstein距离。各种合成和真实世界的数据实验证实了所提出的点过程模型比传统的优越性。

英文摘要:

Point processes are becoming very popular in modeling asynchronous sequential data due to their sound mathematical foundation and strength in modeling a variety of real-world phenomena. Currently, they are often characterized via intensity function which limits model's expressiveness due to unrealistic assumptions on its parametric form used in practice. Furthermore, they are learned via maximum likelihood approach which is prone to failure in multi-modal distributions of sequences. In this paper, we propose an intensity-free approach for point processes modeling that transforms nuisance processes to a target one. Furthermore, we train the model using a likelihood-free leveraging Wasserstein distance between point processes. Experiments on various synthetic and real-world data substantiate the superiority of the proposed point process model over conventional ones.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑