GAN-Based Out-of-Domain Detection Using Both In-Domain and Out-of-Domain Samples

Chaojie Liang; Peijie Huang; Wenbin Lai; Ziheng Ruan

ICASSP 2021accepted0 citations

GAN-Based Out-of-Domain Detection Using Both In-Domain and Out-of-Domain Samples

Chaojie Liang, Peijie Huang, Wenbin Lai, Ziheng Ruan

Abstract

In domain classification for spoken language understanding, correct detection of out-of-domain (OOD) utterances is crucial because it reduces confusion and unnecessary interaction costs between users and the systems. In the situation where both in-domain (ID) and OOD samples are available, our goal is to take advantage of OOD samples under the GAN-based framework for OOD detection. We propose a GAN-based OOD detector with OOD prior distribution and weighted loss (WOODP-GAN). The model consists of a GAN-based detector with OOD prior distribution for generating effective pseudo OOD samples, and a weighted loss function for balancing the loss of fake OOD samples against real OOD samples in the discriminator. Extensive experiments show our proposed WOODP-GAN model outperforms the existing methods in the benchmark dataset CLINC150.

BibTeX

@inproceedings{icassp2021_ganbasedoutofdom,
  title = {GAN-Based Out-of-Domain Detection Using Both In-Domain and Out-of-Domain Samples},
  author = {Chaojie Liang and Peijie Huang and Wenbin Lai and Ziheng Ruan},
  booktitle = {ICASSP 2021},
  year = {2021}
}