Multi-Time Attention Networks for Irregularly Sampled Time Series

Abstract

Irregular sampling occurs in many time series modeling applications where it presents a significant challenge to standard deep learning models. This work is motivated by the analysis of physiological time series data in electronic health records, which are sparse, irregularly sampled, and multivariate. In this paper, we propose a new deep learning framework for this setting that we call Multi-Time Attention Networks. Multi-Time Attention Networks learn an embedding of continuous-time values and use an attention mechanism to produce a fixed-length representation of a time series containing a variable number of observations. We investigate the performance of this framework on interpolation and classification tasks using multiple datasets. Our results show that the proposed approach performs as well or better than a range of baseline and recently proposed models while offering significantly faster training times than current state-of-the-art methods.

arXiv commentAccepted at International Conference on Learning Representations (ICLR) 2021

References in corpus (10)

• Auto-Encoding Variational Bayes
• Empirical Evaluation of Gated Recurrent Neural Networks on Sequence Modeling
• GRU-ODE-Bayes: Continuous modeling of sporadically-observed time series
• Learning to Detect Sepsis with a Multitask Gaussian Process RNN Classifier
• Time2Vec: Learning a Vector Representation of Time
• A scalable end-to-end Gaussian process adapter for irregularly sampled time series classification
• Temporal Self-Attention Network for Medical Concept Embedding
• Self-attention with Functional Time Representation Learning
• Discrete Event, Continuous Time RNNs
• Medical Concept Embedding with Time-Aware Attention

Cited by in corpus (6)

• Graph-Guided Network for Irregularly Sampled Multivariate Time Series
• A Survey on Principles, Models and Methods for Learning from Irregularly Sampled Time Series
• Probabilistic Numeric Convolutional Neural Networks
• Learning Self-Modulating Attention in Continuous Time Space with Applications to Sequential Recommendation
• Multi-axis Attentive Prediction for Sparse EventData: An Application to Crime Prediction
• Towards Generating Real-World Time Series Data