ProtoInfoMax: Prototypical Networks with Mutual Information Maximization for Out-of-Domain Detection

Onderzoeksoutput: Hoofdstuk in Boek/Rapport/CongresprocedureConferentiebijdrageAcademicpeer review

3 Citaten (Scopus)

Samenvatting

The ability to detect Out-of-Domain (OOD) inputs has been a critical requirement in many real-world NLP applications. For example, intent classification in dialogue systems. The reason is that the inclusion of unsupported OOD inputs may lead to catastrophic failure of systems. However, it remains an empirical question whether current methods can tackle such problems reliably in a realistic scenario where zero OOD training data is available. In this study, we propose ProtoInfoMax, a new architecture that extends Prototypical Networks to simultaneously process in-domain and OOD sentences via Mutual Information Maximization (InfoMax) objective. Experimental results show that our proposed method can substantially improve performance up to 20% for OOD detection in low resource settings of text classification. We also show that ProtoInfoMax is less prone to typical overconfidence errors of Neural Networks, leading to more reliable prediction results

Originele taal-2Engels
TitelFindings of the Association for Computational Linguistics, Findings of ACL
SubtitelEMNLP 2021
RedacteurenMarie-Francine Moens, Xuanjing Huang, Lucia Specia, Scott Wen-Tau Yih
Pagina's1606-1617
Aantal pagina's12
ISBN van elektronische versie9781955917100
StatusGepubliceerd - 2021

Bibliografische nota

DBLP License: DBLP's bibliographic metadata records provided through http://dblp.org/ are distributed under a Creative Commons CC0 1.0 Universal Public Domain Dedication. Although the bibliographic metadata records are provided consistent with CC0 1.0 Dedication, the content described by the metadata records is not. Content may be subject to copyright, rights of privacy, rights of publicity and other restrictions.

Vingerafdruk

Duik in de onderzoeksthema's van 'ProtoInfoMax: Prototypical Networks with Mutual Information Maximization for Out-of-Domain Detection'. Samen vormen ze een unieke vingerafdruk.

Citeer dit