Metric Learning and Adaptive Boundary for Out-of-Domain Detection

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F68407700%3A21230%2F22%3A00359318" target="_blank" >RIV/68407700:21230/22:00359318 - isvavai.cz</a>

Nalezeny alternativní kódy

RIV/68407700:21730/22:00359318

Výsledek na webu

<a href="https://doi.org/10.1007/978-3-031-08473-7_12" target="_blank" >https://doi.org/10.1007/978-3-031-08473-7_12</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1007/978-3-031-08473-7_12" target="_blank" >10.1007/978-3-031-08473-7_12</a>

Jazyk výsledku

angličtina

Název v původním jazyce

Metric Learning and Adaptive Boundary for Out-of-Domain Detection

Popis výsledku v původním jazyce

Conversational agents are usually designed for closed-world environments. Unfortunately, users can behave unexpectedly. Based on the open-world environment, we often encounter the situation that the training and test data are sampled from different distributions. Then, data from different distributions are called out-of-domain (OOD). A robust conversational agent needs to react to these OOD utterances adequately. Thus, the importance of robust OOD detection is emphasized. Unfortunately, collecting OOD data is a challenging task. We have designed an OOD detection algorithm independent of OOD data that outperforms a wide range of current state-of-the-art algorithms on publicly available datasets. Our algorithm is based on a simple but efficient approach of combining metric learning with adaptive decision boundary. Furthermore, compared to other algorithms, we have found that our proposed algorithm has significantly improved OOD performance in a scenario with a lower number of classes while preserving the accuracy for in-domain (IND) classes.

Název v anglickém jazyce

Metric Learning and Adaptive Boundary for Out-of-Domain Detection

Popis výsledku anglicky

Conversational agents are usually designed for closed-world environments. Unfortunately, users can behave unexpectedly. Based on the open-world environment, we often encounter the situation that the training and test data are sampled from different distributions. Then, data from different distributions are called out-of-domain (OOD). A robust conversational agent needs to react to these OOD utterances adequately. Thus, the importance of robust OOD detection is emphasized. Unfortunately, collecting OOD data is a challenging task. We have designed an OOD detection algorithm independent of OOD data that outperforms a wide range of current state-of-the-art algorithms on publicly available datasets. Our algorithm is based on a simple but efficient approach of combining metric learning with adaptive decision boundary. Furthermore, compared to other algorithms, we have found that our proposed algorithm has significantly improved OOD performance in a scenario with a lower number of classes while preserving the accuracy for in-domain (IND) classes.

Druh

D - Stať ve sborníku

CEP obor

—

OECD FORD obor

10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)

Projekt

—

Návaznosti

S - Specificky vyzkum na vysokych skolach

Rok uplatnění

2022

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název statě ve sborníku

Natural Language Processing and Information Systems

ISBN

978-3-031-08472-0

ISSN

0302-9743

e-ISSN

—

Počet stran výsledku

8

Strana od-do

127-134

Název nakladatele

Springer, Cham

Místo vydání

—

Místo konání akce

Valencia

Datum konání akce

15. 6. 2022

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

000870296500012