Z-Forcing: Training Stochastic Recurrent Networks

Abstract

Many efforts have been devoted to training generative latent variable models with autoregressive decoders, such as recurrent neural networks (RNN). Stochastic recurrent models have been successful in capturing the variability observed in natural sequential data such as speech. We unify successful ideas from recently proposed architectures into a stochastic recurrent model: each step in the sequence is associated with a latent variable that is used to condition the recurrent dynamics for future steps. Training is performed with amortized variational inference where the approximate posterior is augmented with a RNN that runs backward through the sequence. In addition to maximizing the variational lower bound, we ease training of the latent variables by adding an auxiliary cost which forces them to reconstruct the state of the backward recurrent network. This provides the latent variables with a task-independent objective that enhances the performance of the overall model. We found this strategy to perform better than alternative approaches such as KL annealing. Although being conceptually simple, our model achieves state-of-the-art results on standard speech benchmarks such as TIMIT and Blizzard and competitive performance on sequential MNIST. Finally, we apply our model to language modeling on the IMDB dataset where the auxiliary cost helps in learning interpretable latent variables. Source Code: \url{https://github.com/anirudh9119/zforcing_nips17}

To appear in NIPS'17

References in corpus (12)

• Adam: A Method for Stochastic Optimization
• DRAW: A Recurrent Neural Network For Image Generation
• Modeling Temporal Dependencies in High-Dimensional Sequences: Application to Polyphonic Music Generation and Transcription
• Professor Forcing: A New Algorithm for Training Recurrent Networks
• Variational Lossy Autoencoder
• Learning Stochastic Recurrent Networks
• Learning Discourse-level Diversity for Neural Dialog Models using Conditional Variational Autoencoders
• Sequential Neural Models with Stochastic Layers
• Toward Controlled Generation of Text
• Deep Variational Bayes Filters: Unsupervised Learning of State Space Models from Raw Data
• A Hybrid Convolutional Variational Autoencoder for Text Generation
• Iterative Neural Autoregressive Distribution Estimator (NADE-k)

Cited by in corpus (8)

• Sequential Latent Knowledge Selection for Knowledge-Grounded Dialogue
• Semi-Amortized Variational Autoencoders
• Hierarchical Autoregressive Image Models with Auxiliary Decoders
• Noisin: Unbiased Regularization for Recurrent Neural Networks
• Stochastic WaveNet: A Generative Latent Variable Model for Sequential Data
• Texar: A Modularized, Versatile, and Extensible Toolkit for Text Generation
• Innovations Autoencoder and its Application in One-class Anomalous Sequence Detection
• mu-Forcing: Training Variational Recurrent Autoencoders for Text Generation