Plis Sergey, Danks David, Yang Jianyu
Mind Research Network & University of New Mexico, Albuquerque, NM 87106.
Department of Philosophy, Carnegie Mellon University, Pittsburgh, PA 15213.
Uncertain Artif Intell. 2015 Jul 12;31.
Standard time series structure learning algorithms assume that the measurement timescale is approximately the same as the timescale of the underlying (causal) system. In many scientific contexts, however, this assumption is violated: the measurement timescale can be substantially slower than the system timescale (so intermediate time series datapoints will be missing). This assumption violation can lead to significant learning errors. In this paper, we provide a novel learning algorithm to extract system-timescale structure from measurement data that undersample the underlying system. We employ multiple algorithmic optimizations that exploit the problem structure in order to achieve computational tractability. The resulting algorithm is highly reliable at extracting system-timescale structure from undersampled data.
标准时间序列结构学习算法假定测量时间尺度与潜在（因果）系统的时间尺度大致相同。然而，在许多科学背景下，这一假设并不成立：测量时间尺度可能比系统时间尺度慢得多（因此中间时间序列数据点会缺失）。这种假设不成立可能导致显著的学习误差。在本文中，我们提供了一种新颖的学习算法，用于从对潜在系统进行欠采样的测量数据中提取系统时间尺度结构。我们采用了多种算法优化方法，利用问题结构以实现计算的可处理性。所得算法在从欠采样数据中提取系统时间尺度结构方面具有高度可靠性。
