Timezone: »
Multi-Rate Multivariate Time Series (MR-MTS) are the multivariate time series observations which come with various sampling rates and encode multiple temporal dependencies. State-space models such as Kalman filters and deep learning models such as deep Markov models are mainly designed for time series data with the same sampling rate and cannot capture all the dependencies present in the MR-MTS data. To address this challenge, we propose the Multi-Rate Hierarchical Deep Markov Model (MR-HDMM), a novel deep generative model which uses the latent hierarchical structure with a learnable switch mechanism to capture the temporal dependencies of MR-MTS. Experimental results on two real-world datasets demonstrate that our MR-HDMM model outperforms the existing state-of-the-art deep learning and state-space models on forecasting and interpolation tasks. In addition, the latent hierarchies in our model provide a way to show and interpret the multiple temporal dependencies.
Author Information
Zhengping Che (University of Southern California)
Sanjay Purushotham (University of Southern California)
Max Guangyu Li (University of Southern California)
Bo Jiang (University of Southern California)
Yan Liu (University of Southern California)
Related Events (a corresponding poster, oral, or spotlight)
-
2018 Poster: Hierarchical Deep Generative Models for Multi-Rate Multivariate Time Series »
Fri. Jul 13th 04:15 -- 07:00 PM Room Hall B #54
More from the Same Authors
-
2022 : Estimating Treatment Effects in Continuous Time with Hidden Confounders »
Defu Cao · James Enouen · Yan Liu -
2021 : [06:40 - 07:25 PM UTC] Invited Talk 6: Deciphering Neural Networks through the Lenses of Feature Interactions »
Yan Liu -
2017 Workshop: Time Series Workshop »
Vitaly Kuznetsov · Yan Liu · Scott Yang · Rose Yu -
2017 Tutorial: Deep Learning for Health Care Applications: Challenges and Solutions »
Yan Liu · Jimeng Sun