Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

One-Step Graph-Structured Neural Flows for Irregular Multivariate Time Series Classification

About

Neural Flows efficiently model irregular multivariate time series by directly learning ODE solution trajectories with neural networks, bypassing step-by-step numerical solvers. Despite their efficiency, many existing approaches treat variables independently, leaving inter-variable interactions underexplored. Moreover, their one-step mapping makes interaction modeling inherently challenging, as it removes the iterative refinement of interactions during learning. To address this challenge, we propose one-step Graph-Structured Neural Flows (GSNF), which introduce two auxiliary-trajectory self-supervision strategies to strengthen interaction learning: (i) interaction-aware trajectory generation via re-initialization, which induces trajectory divergence to expose graph-induced interactions, with a theoretically derived lower bound on divergence; and (ii) reverse-time trajectory generation, which enforces forward-backward consistency to regularize graph learning, enabled by flow invertibility. Experiments on five real-world datasets show that GSNF achieves state-of-the-art classification performance with highly competitive training time and memory usage.

Mengzhou Gao, Kaiwei Wang, Pengfei Jiao• 2026

Related benchmarks

TaskDatasetResultRank
Mortality PredictionPhysioNet 2012 (test)
AUC86.7
43
Multivariate Time Series ClassificationPhysionet12
AUROC86.7
15
Multivariate Time Series ClassificationP12
AUROC87.5
15
Multivariate Time Series ClassificationP19
AUROC91.2
15
Multivariate Time Series ClassificationMIMIC IV
AUROC85.3
15
Multivariate Time Series ClassificationeICU
AUROC85.2
15
Showing 6 of 6 rows

Other info

Follow for update