Adversarial Text Generation Without Reinforcement Learning
This addresses the challenge of inefficient GAN training in natural language processing for researchers and practitioners in AI and NLP, though it is incremental as it builds on existing autoencoder and GAN techniques.
The paper tackles the problem of training Generative Adversarial Networks (GANs) for text generation without using reinforcement learning, which is inefficient, by proposing a method that uses an autoencoder to learn a low-dimensional representation of sentences and then trains a GAN in this space, resulting in realistic text generation as shown by human ratings and BLEU scores.
Generative Adversarial Networks (GANs) have experienced a recent surge in popularity, performing competitively in a variety of tasks, especially in computer vision. However, GAN training has shown limited success in natural language processing. This is largely because sequences of text are discrete, and thus gradients cannot propagate from the discriminator to the generator. Recent solutions use reinforcement learning to propagate approximate gradients to the generator, but this is inefficient to train. We propose to utilize an autoencoder to learn a low-dimensional representation of sentences. A GAN is then trained to generate its own vectors in this space, which decode to realistic utterances. We report both random and interpolated samples from the generator. Visualization of sentence vectors indicate our model correctly learns the latent space of the autoencoder. Both human ratings and BLEU scores show that our model generates realistic text against competitive baselines.