arXiv：cs.LG· Perrine Chassat, Agathe Guilloux·· 4 小时前AI 评分33
HyperNSDE：面向静态-纵向临床数据联合生成的个性化神经 SDE
HyperNSDE: Personalized Neural SDEs for Joint Static-Longitudinal Clinical Data Generation
AI 导读
HyperNSDE 是一种连续时间生成模型，通过超网络将潜在 Neural SDE 以静态患者表征为条件，让基线特征塑造轨迹演化而无需轨迹编码器，同时用潜状态依赖的强度过程联合建模观测时间。训练通过确定性-随机路径分解与非对抗性 signature-kernel 目标稳定。在模拟和真实临床数据集上，观测时间保真度提升，预测与相关性指标受观测网格规律性和轨迹平滑度影响。
正文
Abstract:Synthetic patient data generation is a promising solution to the dual challenge of data scarcity and privacy constraints in healthcare machine learning. Realistic synthesis of patient-level clinical data requires jointly modeling heterogeneous static covariates, irregularly sampled longitudinal trajectories, and informative observation times - three tightly coupled components in practice yet rarely addressed together. We propose HyperNSDE, a continuous-time generative model that conditions a latent Neural SDE on static patient representations through a hypernetwork, allowing baseline characteristics to shape trajectory evolution beyond the initial condition without requiring a trajectory encoder, while stochastic latent dynamics capture realistic variability in generated paths. Observation times are modeled jointly through a latent-state-dependent intensity process, and training on irregular stochastic paths is stabilized via a deterministic-stochastic path decomposition with a non-adversarial signature-kernel objective. Experiments on simulated and real clinical datasets show improved observation-time fidelity and competitive performance, while matched-grid analyses reveal that forecasting and correlation metrics are affected by observation-grid regularity and trajectory smoothness.
| Subjects: | Machine Learning (stat.ML); Machine Learning (cs.LG); Applications (stat.AP); Methodology (stat.ME) |
| Cite as: | arXiv:2610.07383 [stat.ML] |
| (or arXiv:2610.07383v1 [stat.ML] for this version) | |
| https://doi.org/10.48550/arXiv.2610.07383 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Perrine Chassat [view email]
[v1]
Mon, 5 Oct 2026 20:54:11 UTC (2,416 KB)
来源：arXiv：cs.LG · arxiv.org