Multi-space Variational Encoder-Decoders for Semi-supervised Labeled Sequence Transduction
arXiv:1704.01691
Abstract
Labeled sequence transduction is a task of transforming one sequence into another sequence that satisfies desiderata specified by a set of labels. In this paper we propose multi-space variational encoder-decoders, a new model for labeled sequence transduction with semi-supervised learning. The generative model can use neural networks to handle both discrete and continuous latent variables to exploit various features of data. Experiments show that our model provides not only a powerful supervised framework but also can effectively take advantage of the unlabeled data. On the SIGMORPHON morphological inflection benchmark, our model outperforms single-model state-of-art results by a large margin for the majority of languages.
Accepted by ACL 2017
References in corpus (11)
- ADADELTA: An Adaptive Learning Rate Method
- Categorical Reparameterization with Gumbel-Softmax
- Tutorial on Variational Autoencoders
- The Concrete Distribution: A Continuous Relaxation of Discrete Random Variables
- Learning Stochastic Recurrent Networks
- Variational Recurrent Auto-Encoders
- Google's Multilingual Neural Machine Translation System: Enabling Zero-Shot Translation
- Controlling Output Length in Neural Encoder-Decoders
- Language as a Latent Variable: Discrete Generative Models for Sentence Compression
- Variational Neural Machine Translation
- Neural Multi-Source Morphological Reinflection
Cited by in corpus (5)
- Toward Controlled Generation of Text
- Transformer-based Conditional Variational Autoencoder for Controllable Story Generation
- Adaptive Self-training for Few-shot Neural Sequence Labeling
- Unsupervised Discrete Sentence Representation Learning for Interpretable Neural Dialog Generation
- An Adversarial Approach to High-Quality, Sentiment-Controlled Neural Dialogue Generation