Poster
Sequence Modeling via Segmentations
Chong Wang · Yining Wang · Po-Sen Huang · Abdelrahman Mohammad · Dengyong Zhou · Li Deng

Tue Aug 8th 06:30 -- 10:00 PM @ Gallery #130

Segmental structure is a common pattern in many types of sequences such as phrases in human languages. In this paper, we present a probabilistic model for sequences via their segmentations. The probability of a segmented sequence is calculated as the product of the probabilities of all its segments, where each segment is modeled using existing tools such as recurrent neural networks. Since the segmentation of a sequence is usually unknown in advance, we sum over all valid segmentations to obtain the final probability for the sequence. An efficient dynamic programming algorithm is developed for forward and backward computations without resorting to any approximation. We demonstrate our approach on text segmentation and speech recognition tasks. In addition to quantitative results, we also show that our approach can discover meaningful segments in their respective application contexts.

Author Information

Chong Wang (Microsoft Research)
Yining Wang (CMU)
Po-Sen Huang (Microsoft Research)
Abdelrahman Mohammad (Microsoft)
Dengyong Zhou (Microsoft Research)
Li Deng (Citadel)

Related Events (a corresponding poster, oral, or spotlight)

More from the Same Authors