Deep Learning for Multivariate Time Series Imputation: A Survey

Published in

International Joint Conference on Artificial In...(2024)

External Links:

Generate Graph

TL;DR

This survey provides a comprehensive summary of deep learning approaches for multivariate time series imputation (MTSI) tasks and proposes a novel taxonomy that categorizes existing methods based on two key perspectives: imputation uncertainty and neural network architecture.

Abstract

Missing values are ubiquitous in multivariate time series (MTS) data, posing significant challenges for accurate analysis and downstream applications. In recent years, deep learning-based methods have successfully handled missing data by leveraging complex temporal dependencies and learned data distributions. In this survey, we provide a comprehensive summary of deep learning approaches for multivariate time series imputation (MTSI) tasks. We propose a novel taxonomy that categorizes existing methods based on two key perspectives: imputation uncertainty and neural network architecture. Furthermore, we summarize existing MTSI toolkits with a particular emphasis on the PyPOTS Ecosystem, which provides an integrated and standardized foundation for MTSI research. Finally, we discuss key challenges and future research directions, which give insight for further MTSI research. This survey aims to serve as a valuable resource for researchers and practitioners in the field of time series analysis and missing data imputation tasks. A well-maintained MTSI paper and tool list is available at https://github.com/WenjieDu/Awesome_Imputation.

Authors

Yuxuan Liang

2 papers

Wenjie Du

3 papers

Jun Wang

1 papers

References74 items

TimeMixer++: A General Time Series Pattern Machine for Universal Predictive Analysis

MTSCI: A Conditional Diffusion Model for Multivariate Time Series Consistent Imputation

How Deep is Your Guess? A Fresh Perspective on Deep Learning for Medical Time-Series Imputation

TSI-Bench: Benchmarking Time Series Imputation

NuwaTS: a Foundation Model Mending Every Incomplete Time Series

Deep Learning for Multivariate Time Series Imputation: A Survey

Published in

International Joint Conference on Artificial In...(2024)

External Links:

Generate Graph

TL;DR

Abstract

Authors

Yuxuan Liang

2 papers

Wenjie Du

3 papers

Jun Wang

1 papers

References74 items

TimeMixer++: A General Time Series Pattern Machine for Universal Predictive Analysis

MTSCI: A Conditional Diffusion Model for Multivariate Time Series Consistent Imputation

How Deep is Your Guess? A Fresh Perspective on Deep Learning for Medical Time-Series Imputation

TSI-Bench: Benchmarking Time Series Imputation

NuwaTS: a Foundation Model Mending Every Incomplete Time Series

Wenjia Wang

1 papers

Qingsong Wen

1 papers

Higher-order Spatio-temporal Physics-incorporated Graph Neural Network for Multivariate Time Series Imputation

SADI: Similarity-Aware Diffusion Model-Based Imputation for Incomplete Temporal EHR Data

A Survey on Diffusion Models for Time Series and Spatio-Temporal Data

Foundation Models for Time Series Analysis: A Tutorial and Survey

Causality-Aware Spatiotemporal Graph Neural Networks for Spatiotemporal Time Series Imputation

Diffusion-TS: Interpretable Diffusion for General Time Series Generation

MOMENT: A Family of Open Time-series Foundation Models

Timer: Generative Pre-trained Transformers Are Large Time Series Models

ImputeFormer: Low Rankness-Induced Transformers for Generalizable Spatiotemporal Imputation

Density-Aware Temporal Attentive Step-wise Diffusion Model For Medical Time Series Imputation

Large Models for Time Series and Spatio-Temporal Data: A Survey and Outlook

Time-LLM: Time Series Forecasting by Reprogramming Large Language Models

Enhancing Missing Data Imputation of Non-stationary Signals with Harmonic Decomposition

Probabilistic Imputation for Time-series Classification with Missing Data

An Observed Value Consistent Diffusion Model for Imputing Missing Values in Multivariate Time Series

Provably Convergent Schrödinger Bridge with Applications to Probabilistic Time Series Imputation

Data Imputation for Sparse Radio Maps in Indoor Positioning

One Fits All: Power General Time Series Analysis by Pretrained LM

PriSTI: A Conditional Diffusion Framework for Spatiotemporal Imputation

Missing Value Imputation for Multi-View Urban Statistical Data via Spatial Correlation Learning

Modeling Temporal Data as Continuous Functions with Stochastic Process Diffusion

TimesNet: Temporal 2D-Variation Modeling for General Time Series Analysis

Diffusion-based Time Series Imputation and Forecasting with Structured State Space Models

Learning to Reconstruct Missing Data from Spatiotemporal Graphs with Sparse Observations

SAITS: Self-Attention-based Imputation for Time Series

Transformers in Time Series: A Survey

RePaint: Inpainting using Denoising Diffusion Probabilistic Models

Differentiable and Scalable Generative Adversarial Models for Data Imputation

MIRACLE: Causally-Aware Imputation via Learning Missing Data Mechanisms

Efficiently Modeling Long Sequences with Structured State Spaces

Filling the G_ap_s: Multivariate Time Series Imputation by Graph Neural Networks

CSDI: Conditional Score-based Diffusion Models for Probabilistic Time Series Imputation

Generative Semi-supervised Learning for Multivariate Time Series Imputation

Missing Value Imputation on Multidimensional Time Series

NRTSI: Non-Recurrent Time Series Imputation

Informer: Beyond Efficient Transformer for Long Sequence Time-Series Forecasting

Time Series Data Imputation: A Survey on Deep Learning Approaches

Uncertainty-Aware Variational-Recurrent Imputation Network for Clinical Time Series

sktime: A Unified Interface for Machine Learning with Time Series

E²GAN: End-to-End Generative Adversarial Network for Multivariate Time Series Imputation

GP-VAE: Deep Probabilistic Time Series Imputation

CDSA: Cross-Dimensional Self-Attention for Multivariate, Geo-tagged Time Series Imputation

Temporal Belief Memory: Imputing Missing Data during RNN Training

BRITS: Bidirectional Recurrent Imputation for Time Series

Estimating Missing Data in Temporal Data Streams Using Multi-Directional Recurrent Neural Networks

Cautionary tales on air-quality improvement in Beijing

Attention is All you Need

Real-valued (Medical) Time Series Generation with Recurrent Conditional GANs

Missing data imputation using fuzzy-rough methods

Recurrent Neural Networks for Multivariate Time Series with Missing Values

Missing data in clinical studies: issues and methods.

Predicting in-hospital mortality of ICU patients: The PhysioNet/Computing in cardiology challenge 2012

MICE: Multivariate Imputation by Chained Equations in R

Improving artificial neural networks' performance in seasonal time series forecasting

Forecasting economic time series using targeted predictors

An Introduction to Kernel and Nearest-Neighbor Nonparametric Regression

INFERENCE AND MISSING DATA

Time series analysis, forecasting and control

Frequency-aware Generative Models for Multivariate Time Series Imputation

Enhance time series modeling by integrating llm

Multivariate Time-series Imputation with Disentangled Temporal Representations

PyPOTS: a Python toolbox for data mining on Partially-Observed Time Series

Variational Inference for Discriminative Learning with Generative Modeling of Feature Incompletion

Mind the Gap: An Experimental Evaluation of Imputation of Missing Values Techniques in Time Series

GluonTS: Probabilistic and Neural Time Series Modeling in Python

Non-autoregressive multiresolution sequence imputation

Statistical analysis with missing data , volume 793

Multivariate Time Series Imputation with Generative Adversarial Networks

imputeTS: Time Series Missing Value Imputation in R

Field of Study

Computer Science

Journal Information

Name

ArXiv

Volume

abs/2005.00687

Venue Information

Name

International Joint Conference on Artificial Intelligence

Type

conference

URL

http://www.ijcai.org/

Alternate Names

Int Jt Conf Artif Intell
IJCAI