Adversarial Feature Matching for Text Generation
arXiv:1706.03850
Abstract
The Generative Adversarial Network (GAN) has achieved great success in generating realistic (real-valued) synthetic data. However, convergence issues and difficulties dealing with discrete data hinder the applicability of GAN to text. We propose a framework for generating realistic text via adversarial training. We employ a long short-term memory network as generator, and a convolutional network as discriminator. Instead of using the standard objective of GAN, we propose matching the high-dimensional latent feature distributions of real and synthetic sentences, via a kernelized discrepancy metric. This eases adversarial training by alleviating the mode-collapsing problem. Our experiments show superior performance in quantitative evaluation, and demonstrate that our model can generate realistic-looking sentences.
Accepted by ICML 2017
References in corpus (10)
- Conditional Generative Adversarial Nets
- Learning Phrase Representations using RNN Encoder-Decoder for Statistical Machine Translation
- Theano: new features and speed improvements
- Convolutional Neural Network Architectures for Matching Natural Language Sentences
- Energy-based Generative Adversarial Network
- A Convolutional Neural Network for Modelling Sentences
- Professor Forcing: A New Algorithm for Training Recurrent Networks
- Aligning Books and Movies: Towards Story-like Visual Explanations by Watching Movies and Reading Books
- Learning to Draw Samples: With Application to Amortized MLE for Generative Adversarial Learning
- On the High-dimensional Power of Linear-time Kernel Two-Sample Testing under Mean-difference Alternatives