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

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Xiao, Jingge, Basso, Leonie, Nejdl, Wolfgang, Ganguly, Niloy, Sikdar, Sandipan
Natura: Preprint
Pubblicazione: 2023
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866916122832404480
author Xiao, Jingge
Basso, Leonie
Nejdl, Wolfgang
Ganguly, Niloy
Sikdar, Sandipan
author_facet Xiao, Jingge
Basso, Leonie
Nejdl, Wolfgang
Ganguly, Niloy
Sikdar, Sandipan
contents 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.
format Preprint
id arxiv_https___arxiv_org_abs_2305_06741
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle IVP-VAE: Modeling EHR Time Series with Initial Value Problem Solvers
Xiao, Jingge
Basso, Leonie
Nejdl, Wolfgang
Ganguly, Niloy
Sikdar, Sandipan
Machine Learning
Artificial Intelligence
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.
title IVP-VAE: Modeling EHR Time Series with Initial Value Problem Solvers
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2305.06741