Projective Latent Dependency Forest Models

Projective Latent Dependency Forest Models
复制标题

投射潜在依赖森林模型

DOI:
10.1109/access.2019.2891292
复制
发表时间:
2019-01
期刊:
影响因子:
3.9
通讯作者:
Kewei Tu
Kewei Tu
中科院分区:
计算机科学3区
文献类型:
--
作者:
Yong Jiang;Yang Zhou;Kewei Tu

文献摘要

参考文献

相似文献

潜在依赖森林模型(LDFM)是一种新型的概率模型,其优点是在模型学习中不需要复杂的结构学习过程。然而,归一化联合概率计算和边际推理是LDFM的难题。在本文中,我们提出了投影LDFM(PLDFM),它是LDFM的一种变体,其联合概率和边缘概率易于计算(关于随机变量数量的立方时间),同时学习仍然容易。我们证明了PLDFM可以看作和积网络(SPN)的一个特例。然后,我们提出了和积投影依赖网络,它是PLDFM和SPN的组合,可以扩展到大量的随机变量。我们在29个数据集上的大量实验表明,我们的模型取得了与其他概率模型相当的结果。
Latent dependence forest models (LDFM) are a new type of probabilistic models with the advantage of not requiring the difficult procedure of structure learning in model learning. However, normalized joint probability computation and marginal inference are intractable for LDFM. In this paper, we proposed projective LDFMs (PLDFMs), a variant of LDFM, for which joint and marginal probabilities become tractable (cubic time with respect to the number of random variables) to compute while learning remains easy. We show that PLDFMs can be seen as a special case of sum-product networks (SPNs). We then propose sum-product projective dependence networks, a combination of PLDFMs and SPNs that scales up to a large number of random variables. Our extensive experiments on 29 datasets show that our models achieve competitive results with other probabilistic models.
DOI: --
发表时间: 2017-03
期刊: ArXiv
影响因子: --
作者:
Diarmaid Conaty;Cassio Polpo de Campos-;D. Mauá
通讯作者: Diarmaid Conaty;Cassio Polpo de Campos-;D. Mauá
DOI: 10.3115/1218955.1219016
发表时间: 2004-07
期刊: --
影响因子: --
作者:
D. Klein;Christopher D. Manning
通讯作者: D. Klein;Christopher D. Manning
DOI: --
发表时间: 2008-12
期刊: --
影响因子: --
作者:
Shay B. Cohen;Kevin Gimpel;Noah A. Smith
通讯作者: Shay B. Cohen;Kevin Gimpel;Noah A. Smith
DOI: 10.1109/iccvw.2011.6130310
发表时间: 2011-07
期刊: 2011 IEEE International Conference on Computer Vision Workshops (ICCV Workshops)
影响因子: --
作者:
Hoifung Poon;Pedro M. Domingos
通讯作者: Hoifung Poon;Pedro M. Domingos
DOI: --
发表时间: 2001-06
期刊: --
影响因子: --
作者:
M. Paskin
通讯作者: M. Paskin