Learning Individual Dynamics from Sparse Cross-Sectional Snapshots

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Lagemann, Christian, Lagemann, Kai, Brunton, Steven L., Mukherjee, Sach
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917523624034304
author Lagemann, Christian
Lagemann, Kai
Brunton, Steven L.
Mukherjee, Sach
author_facet Lagemann, Christian
Lagemann, Kai
Brunton, Steven L.
Mukherjee, Sach
contents Predicting how a dynamical unit evolves over time - how an individual ages, an epidemic spreads, or a physical system degrades - typically requires dense longitudinal tracking. When only extremely sparse or entirely cross-sectional data is available, inferring individualized, continuous-time trajectories is fundamentally ill-posed. Existing methods force a strict compromise: sequence models (e.g. latent ODEs) require dense longitudinal data, while cross-sectional methods (e.g. optimal transport, flow matching-based) map aggregate populations, losing individual dynamics. In this paper, we demonstrate that this dichotomy can be broken. We introduce CADENCE, a principled probabilistic framework that recovers continuous individual trajectories from isolated snapshots by anchoring latent dynamics to static, individual-level contexts. We provide novel identifiability guarantees for single-timepoint trajectory inference. By combining a score-based spatial encoder (bijective Probability Flow ODE) to eliminate diffeomorphic ambiguities with a Soft Mixture-of-Experts (SMoE) router, we show that individual dynamical parameters and routing function are jointly identifiable. Across a suite of benchmarks spanning physical systems to real-world biological data, CADENCE, trained strictly on extremely sparse snapshots with context structure, matches or exceeds the performance of state-of-the-art sequential models trained on dense, full-trajectory data.
format Preprint
id arxiv_https___arxiv_org_abs_2605_23470
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Learning Individual Dynamics from Sparse Cross-Sectional Snapshots
Lagemann, Christian
Lagemann, Kai
Brunton, Steven L.
Mukherjee, Sach
Machine Learning
Artificial Intelligence
Computational Engineering, Finance, and Science
Predicting how a dynamical unit evolves over time - how an individual ages, an epidemic spreads, or a physical system degrades - typically requires dense longitudinal tracking. When only extremely sparse or entirely cross-sectional data is available, inferring individualized, continuous-time trajectories is fundamentally ill-posed. Existing methods force a strict compromise: sequence models (e.g. latent ODEs) require dense longitudinal data, while cross-sectional methods (e.g. optimal transport, flow matching-based) map aggregate populations, losing individual dynamics. In this paper, we demonstrate that this dichotomy can be broken. We introduce CADENCE, a principled probabilistic framework that recovers continuous individual trajectories from isolated snapshots by anchoring latent dynamics to static, individual-level contexts. We provide novel identifiability guarantees for single-timepoint trajectory inference. By combining a score-based spatial encoder (bijective Probability Flow ODE) to eliminate diffeomorphic ambiguities with a Soft Mixture-of-Experts (SMoE) router, we show that individual dynamical parameters and routing function are jointly identifiable. Across a suite of benchmarks spanning physical systems to real-world biological data, CADENCE, trained strictly on extremely sparse snapshots with context structure, matches or exceeds the performance of state-of-the-art sequential models trained on dense, full-trajectory data.
title Learning Individual Dynamics from Sparse Cross-Sectional Snapshots
topic Machine Learning
Artificial Intelligence
Computational Engineering, Finance, and Science
url https://arxiv.org/abs/2605.23470