arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1605.06715stat.MLcs.LG
收藏

基于时间因子的Sigmoid信念网络序列学习

Factored Temporal Sigmoid Belief Networks for Sequence Learning

Tsinghua University(清华大学);Duke University(杜克大学)

Jiaming Song, Zhe Gan, Lawrence Carin

摘要:

深度条件生成模型被开发用于同时学习多个序列的时间依赖性。该模型通过引入三向权张量来捕捉边信息和序列之间的乘法相互作用。所提出的模型建立在时间S形信念网络(TSBN)上,这是S形信念网络(SBN)的顺序堆栈。转移矩阵进一步分解,以减少参数的数量,提高泛化能力。当边信息不可用时,基于所提出的模型的半监督学习的一般框架构成,允许鲁棒的序列分类。实验结果表明,该方法在序列数据上实现了最佳的预测和分类性能,并具有合成序列的能力,具有可控的风格转换和混合。

英文摘要:

Deep conditional generative models are developed to simultaneously learn the temporal dependencies of multiple sequences. The model is designed by introducing a three-way weight tensor to capture the multiplicative interactions between side information and sequences. The proposed model builds on the Temporal Sigmoid Belief Network (TSBN), a sequential stack of Sigmoid Belief Networks (SBNs). The transition matrices are further factored to reduce the number of parameters and improve generalization. When side information is not available, a general framework for semi-supervised learning based on the proposed model is constituted, allowing robust sequence classification. Experimental results show that the proposed approach achieves state-of-the-art predictive and classification performance on sequential data, and has the capacity to synthesize sequences, with controlled style transitioning and blending.

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。

↑