Timezone: »

Normalizing Kalman Filters for Multivariate Time Series Analysis
Emmanuel de Bézenac · Syama Sundar Rangapuram · Konstantinos Benidis · Michael Bohlke-Schneider · Richard Kurle · Lorenzo Stella · Hilaf Hasson · Patrick Gallinari · Tim Januschowski

Wed Dec 09 09:00 AM -- 11:00 AM (PST) @ Poster Session 3 #1789

This paper tackles the modelling of large, complex and multivariate time series panels in a probabilistic setting. To this extent, we present a novel approach reconciling classical state space models with deep learning methods. By augmenting state space models with normalizing flows, we mitigate imprecisions stemming from idealized assumptions in state space models. The resulting model is highly flexible while still retaining many of the attractive properties of state space models, e.g., uncertainty and observation errors are properly accounted for, inference is tractable, sampling is efficient, good generalization performance is observed, even in low data regimes. We demonstrate competitiveness against state-of-the-art deep learning methods on the tasks of forecasting real world data and handling varying levels of missing data.

Author Information

Emmanuel de Bézenac (Sorbonne Université)
Syama Sundar Rangapuram (Amazon Research)
Konstantinos Benidis (Amazon Research)
Michael Bohlke-Schneider (Amazon)
Richard Kurle (Technical University of Munich)
Lorenzo Stella (Amazon Research)
Hilaf Hasson (Amazon Research)
Patrick Gallinari (Sorbonne University & Criteo AI Lab, Paris)
Tim Januschowski (Amazon Research)

- Director Pricing Platform, Zalando SE - Head of Time Series ML at AWS AI

More from the Same Authors