Model Transfer with Explicit Knowledge of the Relation between Class Definitions

Hiyori Yoshikawa, Tomoya Iwakura


Abstract
This paper investigates learning methods for multi-class classification using labeled data for the target classification scheme and another labeled data for a similar but different classification scheme (support scheme). We show that if we have prior knowledge about the relation between support and target classification schemes in the form of a class correspondence table, we can use it to improve the model performance further than the simple multi-task learning approach. Instead of learning the individual classification layers for the support and target schemes, the proposed method converts the class label of each example on the support scheme into a set of candidate class labels on the target scheme via the class correspondence table, and then uses the candidate labels to learn the classification layer for the target scheme. We evaluate the proposed method on two tasks in NLP. The experimental results show that our method effectively learns the target schemes especially for the classes that have a tight connection to certain support classes.
Anthology ID:
K18-1052
Volume:
Proceedings of the 22nd Conference on Computational Natural Language Learning
Month:
October
Year:
2018
Address:
Brussels, Belgium
Editors:
Anna Korhonen, Ivan Titov
Venue:
CoNLL
SIG:
SIGNLL
Publisher:
Association for Computational Linguistics
Note:
Pages:
541–550
Language:
URL:
https://aclanthology.org/K18-1052
DOI:
10.18653/v1/K18-1052
Bibkey:
Cite (ACL):
Hiyori Yoshikawa and Tomoya Iwakura. 2018. Model Transfer with Explicit Knowledge of the Relation between Class Definitions. In Proceedings of the 22nd Conference on Computational Natural Language Learning, pages 541–550, Brussels, Belgium. Association for Computational Linguistics.
Cite (Informal):
Model Transfer with Explicit Knowledge of the Relation between Class Definitions (Yoshikawa & Iwakura, CoNLL 2018)
Copy Citation:
PDF:
https://preview.aclanthology.org/ingest-2024-clasp/K18-1052.pdf
Data
GENIA