ARAML: A Stable Adversarial Training Framework for Text Generation
arXiv:1908.07195
Abstract
Most of the existing generative adversarial networks (GAN) for text generation suffer from the instability of reinforcement learning training algorithms such as policy gradient, leading to unstable performance. To tackle this problem, we propose a novel framework called Adversarial Reward Augmented Maximum Likelihood (ARAML). During adversarial training, the discriminator assigns rewards to samples which are acquired from a stationary distribution near the data rather than the generator's distribution. The generator is optimized with maximum likelihood estimation augmented by the discriminator's rewards instead of policy gradient. Experiments show that our model can outperform state-of-the-art text GANs with a more stable training process.
arXiv commentAccepted by EMNLP 2019
References in corpus (13)
- Sequence to Sequence Learning with Neural Networks
- Microsoft COCO Captions: Data Collection and Evaluation Server
- Scheduled Sampling for Sequence Prediction with Recurrent Neural Networks
- A Neural Conversational Model
- On the Properties of Neural Machine Translation: Encoder-Decoder Approaches
- How NOT To Evaluate Your Dialogue System: An Empirical Study of Unsupervised Evaluation Metrics for Dialogue Response Generation
- MaskGAN: Better Text Generation via Filling in the______
- Topic Aware Neural Response Generation
- Maximum-Likelihood Augmented Discrete Generative Adversarial Networks
- Adversarial Ranking for Language Generation
- RUBER: An Unsupervised Method for Automatic Evaluation of Open-Domain Dialog Systems
- On Accurate Evaluation of GANs for Language Generation
- The price of debiasing automatic metrics in natural language evaluation