TY - GEN
T1 - Networked Time Series Imputation via Position-aware Graph Enhanced Variational Autoencoders
AU - Wang, Dingsu
AU - Yan, Yuchen
AU - Qiu, Ruizhong
AU - Zhu, Yada
AU - Guan, Kaiyu
AU - Margenot, Andrew
AU - Tong, Hanghang
N1 - Publisher Copyright:
© 2023 Owner/Author.
PY - 2023/8/6
Y1 - 2023/8/6
N2 - Multivariate time series (MTS) imputation is a widely studied problem in recent years. Existing methods can be divided into two main groups, including (1) deep recurrent or generative models that primarily focus on time series features, and (2) graph neural networks (GNNs) based models that utilize the topological information from the inherent graph structure of MTS as relational inductive bias for imputation. Nevertheless, these methods either neglect topological information or assume the graph structure is fixed and accurately known. Thus, they fail to fully utilize the graph dynamics for precise imputation in more challenging MTS data such as networked time series (NTS), where the underlying graph is constantly changing and might have missing edges. In this paper, we propose a novel approach to overcome these limitations. First, we define the problem of imputation over NTS which contains missing values in both node time series features and graph structures. Then, we design a new model named PoGeVon which leverages variational autoencoder (VAE) to predict missing values over both node time series features and graph structures. In particular, we propose a new node position embedding based on random walk with restart (RWR) in the encoder with provable higher expressive power compared with message-passing based graph neural networks (GNNs). We further design a decoder with 3-stage predictions from the perspective of multi-task learning to impute missing values in both time series and graph structures reciprocally. Experiment results demonstrate the effectiveness of our model over baselines.
AB - Multivariate time series (MTS) imputation is a widely studied problem in recent years. Existing methods can be divided into two main groups, including (1) deep recurrent or generative models that primarily focus on time series features, and (2) graph neural networks (GNNs) based models that utilize the topological information from the inherent graph structure of MTS as relational inductive bias for imputation. Nevertheless, these methods either neglect topological information or assume the graph structure is fixed and accurately known. Thus, they fail to fully utilize the graph dynamics for precise imputation in more challenging MTS data such as networked time series (NTS), where the underlying graph is constantly changing and might have missing edges. In this paper, we propose a novel approach to overcome these limitations. First, we define the problem of imputation over NTS which contains missing values in both node time series features and graph structures. Then, we design a new model named PoGeVon which leverages variational autoencoder (VAE) to predict missing values over both node time series features and graph structures. In particular, we propose a new node position embedding based on random walk with restart (RWR) in the encoder with provable higher expressive power compared with message-passing based graph neural networks (GNNs). We further design a decoder with 3-stage predictions from the perspective of multi-task learning to impute missing values in both time series and graph structures reciprocally. Experiment results demonstrate the effectiveness of our model over baselines.
KW - imputation
KW - networked time series
KW - node positional embeddings
KW - random walk with restart
KW - variational autoencoders
UR - http://www.scopus.com/inward/record.url?scp=85171346881&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85171346881&partnerID=8YFLogxK
U2 - 10.1145/3580305.3599444
DO - 10.1145/3580305.3599444
M3 - Conference contribution
AN - SCOPUS:85171346881
T3 - Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining
SP - 2256
EP - 2268
BT - KDD 2023 - Proceedings of the 29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining
PB - Association for Computing Machinery
T2 - 29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining, KDD 2023
Y2 - 6 August 2023 through 10 August 2023
ER -