Multi-Time Attention Networks for Irregularly Sampled Time Series

TL;DR

This paper introduces Multi-Time Attention Networks, a deep learning framework designed to effectively handle irregularly sampled multivariate time series, especially in healthcare data, achieving competitive accuracy with faster training.

Contribution

The paper presents a novel attention-based model that embeds continuous-time values and efficiently manages irregular sampling in multivariate time series.

Findings

Performs as well or better than existing models on interpolation and classification tasks.

Offers significantly faster training times compared to state-of-the-art methods.

Effective on multiple healthcare-related datasets.

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.