English

IVP-VAE: Modeling EHR Time Series with Initial Value Problem Solvers

Machine Learning 2024-02-14 v3 Artificial Intelligence

Abstract

Continuous-time models such as Neural ODEs and Neural Flows have shown promising results in analyzing irregularly sampled time series frequently encountered in electronic health records. Based on these models, time series are typically processed with a hybrid of an initial value problem (IVP) solver and a recurrent neural network within the variational autoencoder architecture. Sequentially solving IVPs makes such models computationally less efficient. In this paper, we propose to model time series purely with continuous processes whose state evolution can be approximated directly by IVPs. This eliminates the need for recurrent computation and enables multiple states to evolve in parallel. We further fuse the encoder and decoder with one IVP solver utilizing its invertibility, which leads to fewer parameters and faster convergence. Experiments on three real-world datasets show that the proposed method can systematically outperform its predecessors, achieve state-of-the-art results, and have significant advantages in terms of data efficiency.

Keywords

Cite

@article{arxiv.2305.06741,
  title  = {IVP-VAE: Modeling EHR Time Series with Initial Value Problem Solvers},
  author = {Jingge Xiao and Leonie Basso and Wolfgang Nejdl and Niloy Ganguly and Sandipan Sikdar},
  journal= {arXiv preprint arXiv:2305.06741},
  year   = {2024}
}

Comments

AAAI 2024 Camera-Ready Version

R2 v1 2026-06-28T10:31:56.830Z