Smooth dynamics interrupted by discontinuities are known as hybrid systems and arise commonly in nature. Latent ODEs allow for powerful representation of irregularly sampled time series but are not designed to capture trajectories arising from hybrid systems. Here, we propose the Latent Segmented ODE (LatSegODE), which uses Latent ODEs to perform reconstruction and changepoint detection within hybrid trajectories featuring jump discontinuities and switching dynamical modes. Where it is possible to train a Latent ODE on the smooth dynamical flows between discontinuities, we apply the pruned exact linear time (PELT) algorithm to detect changepoints where latent dynamics restart, thereby maximizing the joint probability of a piece-wise continuous latent dynamical representation. We propose usage of the marginal likelihood as a score function for PELT, circumventing the need for model complexity-based penalization. The LatSegODE outperforms baselines in reconstructive and segmentation tasks including synthetic data sets of sine waves, Lotka Volterra dynamics, and UCI Character Trajectories.
翻译:因不连续中断而中断的滑动动态被称作混合系统, 通常在性质上产生。 远程代码允许不规则抽样的时间序列有强大的代表性, 但没有设计来捕捉混合系统产生的轨迹。 在这里, 我们建议使用中端分割的 ODE (LatSegode), 使用中端分割的 ODE (LatSegode) 在混合轨迹内进行重建和变化点检测, 包括跳跃不连续和转换动态模式。 如果有可能在不连续的平稳动态流上对中端运行的 ODE 进行培训, 我们则使用经剪切的精确线性时间算法来探测潜在动态重新启动的变更点, 从而最大限度地增加片段连续潜在动态代表的共概率。 我们提议使用边际概率作为 PELT 的计分数函数, 绕开基于模型复杂性的处罚的需要。 LatSegOde 将重组和分化任务的基线( 包括正弦波合成数据集、 Lotka Volterravila 动态和 UCI 直径) 。