Sökning: onr:"swepub:oai:DiVA.org:uu-457404" >
Bidirectional Domai...
Bidirectional Domain Adaptation Using Weighted Multi-Task Learning
-
- Dakota, Daniel (författare)
- Uppsala universitet,Institutionen för lingvistik och filologi
-
- Sayyed, Zeeshan Ali (författare)
- Indiana Univ, Bloomington, IN 47405 USA.
-
- Kuebler, Sandra (författare)
- Indiana Univ, Bloomington, IN 47405 USA.
-
(creator_code:org_t)
- Stroudsburg, PA, USA : Association for Computational Linguistics, 2021
- 2021
- Engelska.
-
Ingår i: IWPT 2021. - Stroudsburg, PA, USA : Association for Computational Linguistics. - 9781954085800 ; , s. 93-105
- Relaterad länk:
-
https://aclanthology...
-
visa fler...
-
https://urn.kb.se/re...
-
https://doi.org/10.1...
-
visa färre...
Abstract
Ämnesord
Stäng
- Domain adaption in syntactic parsing is still a significant challenge. We address the issue of data imbalance between the in-domain and out-of-domain treebank typically used for the problem. We define domain adaptation as a Multi-task learning (MTL) problem, which allows us to train two parsers, one for each domain. Our results show that the MTL approach is beneficial for the smaller treebank. For the larger treebank, we need to use loss weighting in order to avoid a decrease in performance below the single task. In order to determine to what degree the data imbalance between two domains and the domain differences affect results, we also carry out an experiment with two imbalanced in-domain treebanks and show that loss weighting also improves performance in an in-domain setting. Given loss weighting in MTL, we can improve results for both parsers.
Ämnesord
- NATURVETENSKAP -- Data- och informationsvetenskap -- Språkteknologi (hsv//swe)
- NATURAL SCIENCES -- Computer and Information Sciences -- Language Technology (hsv//eng)
- NATURVETENSKAP -- Data- och informationsvetenskap -- Datavetenskap (hsv//swe)
- NATURAL SCIENCES -- Computer and Information Sciences -- Computer Sciences (hsv//eng)
Publikations- och innehållstyp
- ref (ämneskategori)
- kon (ämneskategori)
Hitta via bibliotek
Till lärosätets databas