Dynamic Sum Product Networks for Tractable Inference on Sequence Data
Dynamic Sum Product Networks for Tractable Inference on Sequence Data
复制标题
用于序列数据易处理推理的动态和积网络
DOI：
--
复制
发表时间：
2015
期刊：
影响因子：
--
通讯作者：
George Trimponias
中科院分区：
文献类型：
--
作者：
Mazen Melibari;P. Poupart;Prashant Doshi;George Trimponias
和积网络（SPN）最近作为一类新的易于处理的概率图形模型出现。与贝叶斯网络和马尔可夫网络的推理可能与网络大小呈指数关系不同，SPN 中的推理在时间上与网络大小呈线性关系。由于 SPN 仅表示一组固定变量的分布，因此我们提出动态和积网络 (DSPN) 作为 SPN 对于不同长度的序列数据的推广。 DSPN 由模板网络组成，该模板网络根据需要重复多次，以对任意长度的数据序列进行建模。我们提出了一种本地搜索技术来学习模板网络的结构。与动态贝叶斯网络相比，动态贝叶斯网络的推理通常在每个时间片的变量数量上呈指数增长，DSPN 继承了 SPN 的线性推理复杂性。我们在多个序列数据集上展示了 DSPN 相对于 DBN 和其他模型的优势。
Sum-Product Networks (SPN) have recently emerged as a new class of tractable probabilistic graphical models. Unlike Bayesian networks and Markov networks where inference may be exponential in the size of the network, inference in SPNs is in time linear in the size of the network. Since SPNs represent distributions over a fixed set of variables only, we propose dynamic sum product networks (DSPNs) as a generalization of SPNs for sequence data of varying length. A DSPN consists of a template network that is repeated as many times as needed to model data sequences of any length. We present a local search technique to learn the structure of the template network. In contrast to dynamic Bayesian networks for which inference is generally exponential in the number of variables per time slice, DSPNs inherit the linear inference complexity of SPNs. We demonstrate the advantages of DSPNs over DBNs and other models on several datasets of sequence data.