BRITS: Bidirectional Recurrent Imputation for Time Series

Abstract

Time series are widely used as signals in many classification/regression tasks. It is ubiquitous that time series contains many missing values. Given multiple correlated time series data, how to fill in missing values and to predict their class labels? Existing imputation methods often impose strong assumptions of the underlying data generating process, such as linear dynamics in the state space. In this paper, we propose BRITS, a novel method based on recurrent neural networks for missing value imputation in time series data. Our proposed method directly learns the missing values in a bidirectional recurrent dynamical system, without any specific assumption. The imputed values are treated as variables of RNN graph and can be effectively updated during the backpropagation.BRITS has three advantages: (a) it can handle multiple correlated missing values in time series; (b) it generalizes to time series with nonlinear dynamics underlying; (c) it provides a data-driven imputation procedure and applies to general settings with missing data.We evaluate our model on three real-world datasets, including an air quality dataset, a health-care data, and a localization data for human activity. Experiments show that our model outperforms the state-of-the-art methods in both imputation and classification/regression accuracies.

References in corpus (2)

• Convolutional LSTM Network: A Machine Learning Approach for Precipitation Nowcasting
• Scheduled Sampling for Sequence Prediction with Recurrent Neural Networks

Cited by in corpus (36)

• Adaptive Graph Convolutional Recurrent Network for Traffic Forecasting
• Bayesian Temporal Factorization for Multidimensional Time Series Prediction
• Neural Controlled Differential Equations for Irregular Time Series
• Dynamic Spatiotemporal Graph Convolutional Neural Networks for Traffic Data Imputation with Complex Missing Patterns
• Learning Long-Term Dependencies in Irregularly-Sampled Time Series
• CDSA: Cross-Dimensional Self-Attention for Multivariate, Geo-tagged Time Series Imputation
• STING: Self-attention based Time-series Imputation Networks using GAN
• Data-driven Reconstruction of Nonlinear Dynamics from Sparse Observation
• Networked Time Series Imputation via Position-aware Graph Enhanced Variational Autoencoders
• GP-VAE: Deep Probabilistic Time Series Imputation
• Wind energy forecasting with missing values within a fully conditional specification framework
• NAOMI: Non-Autoregressive Multiresolution Sequence Imputation
• RDIS: Random Drop Imputation with Self-Training for Incomplete Time Series Data
• Why Not to Use Zero Imputation? Correcting Sparsity Bias in Training Neural Networks
• Joint Modeling of Local and Global Temporal Dynamics for Multivariate Time Series Forecasting with Missing Values
• Ball Trajectory Inference from Multi-Agent Sports Contexts Using Set Transformer and Hierarchical Bi-LSTM
• Are deep learning models superior for missing data imputation in large surveys? Evidence from an empirical comparison
• Temporal Graph Signal Decomposition
• Forecasting Loss of Signal in Optical Networks with Machine Learning
• Missing Data as Augmentation in the Earth Observation Domain: A Multi-View Learning Approach
• LIFE: Learning Individual Features for Multivariate Time Series Prediction with Missing Values
• An Empirical Study: Extensive Deep Temporal Point Process
• Seq2Tens: An Efficient Representation of Sequences by Low-Rank Tensor Projections
• Multilevel Stochastic Optimization for Imputation in Massive Medical Data Records
• Physics-aware Spatiotemporal Modules with Auxiliary Tasks for Meta-Learning
• Gradient Importance Learning for Incomplete Observations
• Time Series Continuous Modeling for Imputation and Forecasting with Implicit Neural Representations
• Learning Disentangled Representations of Video with Missing Data
• Time-Series Imputation with Wasserstein Interpolation for Optimal Look-Ahead-Bias and Variance Tradeoff
• Networked Time Series Prediction with Incomplete Data via Generative Adversarial Network
• CoSTI: Consistency Models for (a faster) Spatio-Temporal Imputation
• An Analysis of Temporal Dropout in Earth Observation Time Series for Regression Tasks
• Medical data wrangling with sequential variational autoencoders
• As easy as APC: overcoming missing data and class imbalance in time series with self-supervised learning
• Yformer: U-Net Inspired Transformer Architecture for Far Horizon Time Series Forecasting
• Towards Generating Real-World Time Series Data