Adversarial Feature Matching for Text Generation
This addresses the challenge of applying GANs to text generation for natural language processing tasks, though it appears incremental as it builds on existing GAN methods with a novel feature matching approach.
The paper tackles the problem of generating realistic text using Generative Adversarial Networks (GANs), which face convergence issues and difficulties with discrete data, by proposing a framework that matches high-dimensional latent feature distributions of real and synthetic sentences using a kernelized discrepancy metric, resulting in superior quantitative performance and realistic sentence generation.
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.