We study the problem of learning a mixture of multiple linear dynamical systems (LDSs) from unlabeled short sample trajectories, each generated by one of the LDS models. Despite the wide applicability of mixture models for time-series data, learning algorithms that come with end-to-end performance guarantees are largely absent from existing literature. There are multiple sources of technical challenges, including but not limited to (1) the presence of latent variables (i.e. the unknown labels of trajectories); (2) the possibility that the sample trajectories might have lengths much smaller than the dimension $d$ of the LDS models; and (3) the complicated temporal dependence inherent to time-series data. To tackle these challenges, we develop a two-stage meta-algorithm, which is guaranteed to efficiently recover each ground-truth LDS model up to error $\tilde{O}(\sqrt{d/T})$, where $T$ is the total sample size. We validate our theoretical studies with numerical experiments, confirming the efficacy of the proposed algorithm.
翻译:我们研究从一个LDS模型产生的未经标记的短试样轨迹中学习多种线性动态系统(LDS)的混合物的问题。尽管混合模型对时间序列数据具有广泛适用性,但现有文献中基本上没有包含端到端的性能保障的学习算法。技术挑战的多种来源包括但不局限于:(1) 潜在变量的存在(即轨迹的未知标签);(2) 样本轨迹的长度可能大大小于LDS模型的维度(美元);(3) 时间序列数据固有的复杂时间依赖性。为了应对这些挑战,我们开发了两阶段的元-负数模型,保证有效地回收每个地面-梯度模型,直至出现错误 $\tilde{O}(sqrt{d/T}$,其中美元是总样本大小。我们用数字实验来验证我们的理论研究,证实了拟议的算法的有效性。