History Marginalization Improves Forecasting in Variational Recurrent Neural Networks

Deep probabilistic time series forecasting models have become an integral part of machine learning. While several powerful generative models have been proposed, we provide evidence that their associated inference models are oftentimes too limited and cause the generative model to predict mode-averaged dynamics. Mode-averaging is problematic since many real-world sequences are highly multi-modal, and their averaged dynamics are unphysical (e.g., predicted taxi trajectories might run through buildings on the street map). To better capture multi-modality, we develop variational dynamic mixtures (VDM): a new variational family to infer sequential latent variables. The VDM approximate posterior at each time step is a mixture density network, whose parameters come from propagating multiple samples through a recurrent architecture. This results in an expressive multi-modal posterior approximation. In an empirical study, we show that VDM outperforms competing approaches on highly multi-modal datasets from different domains.

Automated summary

The importance of deep probabilistic time series forecasting models is highlighted in the article, pointing out that existing generative models' inference models are often too limited, resulting in overly averaged dynamics in predictions. A variational dynamic mixtures (VDM) model is developed to address this issue by capturing multi-modality. Empirical studies show that VDM outperforms other methods in handling highly multi-modal datasets.