Saved in:
Bibliographic Details
Main Authors: Liu, Ying, Cui, Peng, Hu, Wenbo, Hong, Richang
Format: Preprint
Published: 2023
Subjects:
Online Access:https://arxiv.org/abs/2312.01294
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911295536627712
author Liu, Ying
Cui, Peng
Hu, Wenbo
Hong, Richang
author_facet Liu, Ying
Cui, Peng
Hu, Wenbo
Hong, Richang
contents Real-world time series data often exhibits substantial missing values, posing challenges for advanced analysis. A common approach to addressing this issue is imputation, where the primary challenge lies in determining the appropriate values to fill in. While previous deep learning methods have proven effective for time series imputation, they often produce overconfident imputations, which poses a potentially overlooked risk to the reliability of the intelligent system. Diffusion methods are proficient in estimating probability distributions but face challenges under a high missing rate and are, moreover, computationally expensive due to the nature of the generative model framework. In this paper, we propose Quantile Sub-Ensembles, a novel method that estimates uncertainty with ensembles of quantile-regression-based task networks and incorporate Quantile Sub-Ensembles into a non-generative time series imputation method. Our method not only produces accurate and reliable imputations, but also remains computationally efficient due to its non-generative framework. We conduct extensive experiments on five real-world datasets, and the results demonstrates superior performance in both deterministic and probabilistic imputation compared to baselines across most experimental settings. The code is available at https://github.com/yingliu-coder/QSE.
format Preprint
id arxiv_https___arxiv_org_abs_2312_01294
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Deep sub-ensembles meets quantile regression: uncertainty-aware imputation for time series
Liu, Ying
Cui, Peng
Hu, Wenbo
Hong, Richang
Machine Learning
Real-world time series data often exhibits substantial missing values, posing challenges for advanced analysis. A common approach to addressing this issue is imputation, where the primary challenge lies in determining the appropriate values to fill in. While previous deep learning methods have proven effective for time series imputation, they often produce overconfident imputations, which poses a potentially overlooked risk to the reliability of the intelligent system. Diffusion methods are proficient in estimating probability distributions but face challenges under a high missing rate and are, moreover, computationally expensive due to the nature of the generative model framework. In this paper, we propose Quantile Sub-Ensembles, a novel method that estimates uncertainty with ensembles of quantile-regression-based task networks and incorporate Quantile Sub-Ensembles into a non-generative time series imputation method. Our method not only produces accurate and reliable imputations, but also remains computationally efficient due to its non-generative framework. We conduct extensive experiments on five real-world datasets, and the results demonstrates superior performance in both deterministic and probabilistic imputation compared to baselines across most experimental settings. The code is available at https://github.com/yingliu-coder/QSE.
title Deep sub-ensembles meets quantile regression: uncertainty-aware imputation for time series
topic Machine Learning
url https://arxiv.org/abs/2312.01294