用轨迹流匹配训练神经随机微分方程，提升临床时间序列建模效果。
Trajectory Flow Matching with Applications to Clinical Time Series Modeling
- 通过流匹配技术无须反向传播求解SDE，实现高效训练
- 在三个临床数据集上提升性能，尤其改善不确定性预测
- 适合处理不规则采样、具有随机性的医疗时间序列
建模随机且不规则采样的时间序列是众多应用中的挑战性问题，尤其在医学领域。神经随机微分方程（Neural SDEs）是一种有吸引力的建模方法，通过神经网络参数化SDE的漂移项和扩散项。然而，现有训练算法需对SDE动力学进行反向传播，严重限制其可扩展性和稳定性。为此，我们提出轨迹流匹配（Trajectory Flow Matching, TFM），以无需仿真方式训练神经SDE，绕过对动态过程的反向传播。TFM借鉴生成建模中的流匹配技术来建模时间序列。本文首先建立了TFM学习时间序列数据的必要条件；其次提出一种重参数化技巧以提升训练稳定性；最后将TFM适配至临床时间序列场景，在三个临床时间序列数据集上均取得更优表现，不仅绝对性能提升，且在不确定性预测方面也更佳。
原文摘要 · Abstract (English)
Modeling stochastic and irregularly sampled time series is a challenging problem found in a wide range of applications, especially in medicine. Neural stochastic differential equations (Neural SDEs) are an attractive modeling technique for this problem, which parameterize the drift and diffusion terms of an SDE with neural networks. However, current algorithms for training Neural SDEs require backpropagation through the SDE dynamics, greatly limiting their scalability and stability. To address this, we propose Trajectory Flow Matching (TFM), which trains a Neural SDE in a simulation-free manner, bypassing backpropagation through the dynamics. TFM leverages the flow matching technique from generative modeling to model time series. In this work we first establish necessary conditions for TFM to learn time series data. Next, we present a reparameterization trick which improves training stability. Finally, we adapt TFM to the clinical time series setting, demonstrating improved performance on three clinical time series datasets both in terms of absolute performance and uncertainty prediction.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。