English

OodGAN: Generative Adversarial Network for Out-of-Domain Data Generation

Computation and Language 2021-04-07 v1 Artificial Intelligence Machine Learning

Abstract

Detecting an Out-of-Domain (OOD) utterance is crucial for a robust dialog system. Most dialog systems are trained on a pool of annotated OOD data to achieve this goal. However, collecting the annotated OOD data for a given domain is an expensive process. To mitigate this issue, previous works have proposed generative adversarial networks (GAN) based models to generate OOD data for a given domain automatically. However, these proposed models do not work directly with the text. They work with the text's latent space instead, enforcing these models to include components responsible for encoding text into latent space and decoding it back, such as auto-encoder. These components increase the model complexity, making it difficult to train. We propose OodGAN, a sequential generative adversarial network (SeqGAN) based model for OOD data generation. Our proposed model works directly on the text and hence eliminates the need to include an auto-encoder. OOD data generated using OodGAN model outperforms state-of-the-art in OOD detection metrics for ROSTD (67% relative improvement in FPR 0.95) and OSQ datasets (28% relative improvement in FPR 0.95) (Zheng et al., 2020).

Keywords

Cite

@article{arxiv.2104.02484,
  title  = {OodGAN: Generative Adversarial Network for Out-of-Domain Data Generation},
  author = {Petr Marek and Vishal Ishwar Naik and Vincent Auvray and Anuj Goyal},
  journal= {arXiv preprint arXiv:2104.02484},
  year   = {2021}
}

Comments

NAACL 2021 Industry track

R2 v1 2026-06-24T00:53:10.865Z