论文标题
隐藏的马尔可夫非线性ICA：非组织时间序列的无监督学习
Hidden Markov Nonlinear ICA: Unsupervised Learning from Nonstationary Time Series
论文作者
论文摘要
非线性独立组件分析（ICA）的最新进展为无监督的特征学习和分离提供了原则上的框架。在此类工作中的主要思想是，在某些观察到的辅助变量（例如时间段指数）上，假定潜在组件是独立的条件。这需要将数据进行手动分割成非平稳的段，这些段在计算上昂贵，不准确且通常是不可能的。因此，这些模型并未完全不受监督。我们通过将非线性ICA与隐藏的Markov模型相结合，从而纠正这些局限性，从而导致潜在状态代替观察到的段索引。我们证明了针对一般混合非线性（例如神经网络）所提出的模型的可识别性。我们还展示了如何使用预期最大化算法完成模型的最大似然估计。因此，我们实现了一个新的非线性ICA框架，该框架是无监督，更有效的，并且能够建模基本的时间动力学。
Recent advances in nonlinear Independent Component Analysis (ICA) provide a principled framework for unsupervised feature learning and disentanglement. The central idea in such works is that the latent components are assumed to be independent conditional on some observed auxiliary variables, such as the time-segment index. This requires manual segmentation of data into non-stationary segments which is computationally expensive, inaccurate and often impossible. These models are thus not fully unsupervised. We remedy these limitations by combining nonlinear ICA with a Hidden Markov Model, resulting in a model where a latent state acts in place of the observed segment-index. We prove identifiability of the proposed model for a general mixing nonlinearity, such as a neural network. We also show how maximum likelihood estimation of the model can be done using the expectation-maximization algorithm. Thus, we achieve a new nonlinear ICA framework which is unsupervised, more efficient, as well as able to model underlying temporal dynamics.
