Generative Adversarial Nets for Multiple Text Corpora
TL
TL;DR
This work demonstrates the GAN models on real-world text data sets from different corpora, and shows that embeddings from both models lead to improvements in supervised learning problems.
Abstract
Generative adversarial nets (GANs) have been successfully applied to the artificial generation of image data. In terms of text data, much has been done on the artificial generation of natural language from a single corpus. We consider multiple text corpora as the input data, for which there can be two applications of GANs: (1) the creation of consistent cross-corpus word embeddings given different word embeddings per corpus; (2) the generation of robust bag-of-words document embeddings for each corpora. We demonstrate our GAN models on real-world text data sets from different corpora, and show that embeddings from both models (weGAN and deGAN) lead to improvements in supervised learning problems.
References27 items
1
Generative Adversarial Networks
2
An End-to-End Generative Architecture for Paraphrase Generation
3
Conditional Generative Adversarial Networks for Commonsense Machine Comprehension
4
Language Generation with Recurrent Generative Adversarial Networks without Pre-training
5