Abstract | ||
---|---|---|
Deep dynamic generative models are developed to learn sequential dependencies in time-series data. The multi-layered model is designed by constructing a hierarchy of temporal sigmoid belief networks (TSBNs), defined as a sequential stack of sigmoid belief networks (SBNs). Each SBN has a contextual hidden state, inherited from the previous SBNs in the sequence, and is used to regulate its hidden bias. Scalable learning and inference algorithms are derived by introducing a recognition model that yields fast sampling from the variational posterior. This recognition model is trained jointly with the generative model, by maximizing its variational lower bound on the log-likelihood. Experimental results on bouncing balls, polyphonic music, motion capture, and text streams show that the proposed approach achieves state-of-the-art predictive performance, and has the capacity to synthesize various sequences. |
Year | Venue | DocType |
---|---|---|
2015 | Annual Conference on Neural Information Processing Systems | Journal |
Volume | ISSN | Citations |
abs/1509.07087 | 1049-5258 | 18 |
PageRank | References | Authors |
0.82 | 23 | 5 |
Name | Order | Citations | PageRank |
---|---|---|---|
Zhe Gan | 1 | 319 | 32.58 |
Chunyuan Li | 2 | 467 | 33.86 |
Ricardo Henao | 3 | 286 | 23.85 |
David E. Carlson | 4 | 182 | 15.35 |
L. Carin | 5 | 4603 | 339.36 |