Deep Learning for Multivariate Time Series Imputation: A Survey

2024-02-06

arXiv.org

AI Summary

Multivariate time series (MTS) imputation faces dual challenges: modeling uncertainty in missing data estimation and architectural diversity across methods. To address these, this work introduces the first two-dimensional taxonomy jointly characterizing imputation uncertainty and network architecture, enabling systematic method categorization. We propose PyPOTS—a unified, open-source framework for MTS imputation research and development—integrating over 100 state-of-the-art models, including RNNs, Transformers, GANs, VAEs, and graph neural networks, augmented with self-supervised learning, adversarial training, and probabilistic modeling. PyPOTS features standardized APIs, reproducible benchmarks, and modular design to streamline method comparison, implementation, and deployment. Empirical evaluation demonstrates that PyPOTS significantly enhances reproducibility, standardization, and efficiency in MTS imputation research, establishing itself as the de facto open-source toolkit in the community.

Technology Category

• Machine Learning: Time-Series/Data Streams
• Reasoning under Uncertainty: Relational Probabilistic Models
• Cognitive Modeling & Cognitive Systems: Neural Spike Coding

Application Category

• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Web Mining and Content Analysis: Robustness and generalizability of Web mining methods

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.

Problem

• Address missing values in multivariate time series data.
• Summarize deep learning approaches for time series imputation.
• Propose taxonomy based on imputation uncertainty and network architecture.

Innovation

• Deep learning for time series imputation
• Taxonomy based on uncertainty and architecture
• PyPOTS Ecosystem for standardized research