Modeling documents with Generative Adversarial Networks
This work addresses document representation learning for natural language processing, but it is incremental as it adapts existing GAN methods.
The paper tackled the problem of learning distributed representations for natural language documents by proposing a model based on Energy-Based GANs with a Denoising Autoencoder as the discriminator, and it evaluated the extracted representations quantitatively and qualitatively.
This paper describes a method for using Generative Adversarial Networks to learn distributed representations of natural language documents. We propose a model that is based on the recently proposed Energy-Based GAN, but instead uses a Denoising Autoencoder as the discriminator network. Document representations are extracted from the hidden layer of the discriminator and evaluated both quantitatively and qualitatively.