We introduce a framework for unsupervised learning of structured predictors with overlapping, global features. Each input's latent representation is predicted conditional on the observable data using a feature-rich conditional random field. Then a reconstruction of the input is (re)generated, conditional on the latent structure, using models for which maximum likelihood estimation has a closed-form. Our autoencoder formulation enables efficient learning without making unrealistic independence assumptions or restricting the kinds of features that can be used. We illustrate insightful connections to traditional autoencoders, posterior regularization and multi-view learning. We show competitive results with instantiations of the model for two canonical NLP tasks: part-of-speech induction and bitext word alignment, and show that training our model can be substantially more efficient than comparable feature-rich baselines.

本文介绍了一个用于重叠的全局特征的无监督学习结构化预测的框架，提出了一个基于特征丰富的条件随机场对可观察数据进行条件预测的潜在表示方法，并使用具有封闭形式的最大似然估计的模型来（重新）生成输入的重建，使得无需进行不切实际的独立性假设或限制可用特征类型的情况下能够有效地进行学习，并示范了与传统自动编码器、后验正则化和多视图学习方面的有见地的联系。接着将模型实例化应用于两个经典的自然语言处理任务：词性归纳和比对文本词汇对应，并证明训练我们的模型可以比可比的特征丰富基线方法更加高效。

无监督结构预测的条件随机场自编码器