Factored temporal sigmoid belief networks for sequence learning

Published

Conference Paper

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-theart predictive and classification performance on sequential data, and has the capacity to synthesize sequences, with controlled style transitioning and blending.

Duke Authors

Cited Authors

  • Song, J; Gan, Z; Carin, L

Published Date

  • January 1, 2016

Published In

  • 33rd International Conference on Machine Learning, Icml 2016

Volume / Issue

  • 3 /

Start / End Page

  • 1937 - 1946

International Standard Book Number 13 (ISBN-13)

  • 9781510829008

Citation Source

  • Scopus