Structured Inference Networks for Nonlinear State Space Models

30 Sep 2016  ·  Rahul G. Krishnan, Uri Shalit, David Sontag ·

Gaussian state space models have been used for decades as generative models of sequential data. They admit an intuitive probabilistic interpretation, have a simple functional form, and enjoy widespread adoption. We introduce a unified algorithm to efficiently learn a broad class of linear and non-linear state space models, including variants where the emission and transition distributions are modeled by deep neural networks. Our learning algorithm simultaneously learns a compiled inference network and the generative model, leveraging a structured variational approximation parameterized by recurrent neural networks to mimic the posterior distribution. We apply the learning algorithm to both synthetic and real-world datasets, demonstrating its scalability and versatility. We find that using the structured approximation to the posterior results in models with significantly higher held-out likelihood.

PDF Abstract


Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Multivariate Time Series Forecasting USHCN-Daily Sequential VAE MSE 0.83 # 5


No methods listed for this paper. Add relevant methods here