Distributional Modeling on a Diet: One-shot Word Learning from Text Only

Su Wang, Stephen Roller, Katrin Erk


Abstract
We test whether distributional models can do one-shot learning of definitional properties from text only. Using Bayesian models, we find that first learning overarching structure in the known data, regularities in textual contexts and in properties, helps one-shot learning, and that individual context items can be highly informative.
Anthology ID:
I17-1021
Volume:
Proceedings of the Eighth International Joint Conference on Natural Language Processing (Volume 1: Long Papers)
Month:
November
Year:
2017
Address:
Taipei, Taiwan
Editors:
Greg Kondrak, Taro Watanabe
Venue:
IJCNLP
SIG:
Publisher:
Asian Federation of Natural Language Processing
Note:
Pages:
204–213
Language:
URL:
https://aclanthology.org/I17-1021
DOI:
Bibkey:
Cite (ACL):
Su Wang, Stephen Roller, and Katrin Erk. 2017. Distributional Modeling on a Diet: One-shot Word Learning from Text Only. In Proceedings of the Eighth International Joint Conference on Natural Language Processing (Volume 1: Long Papers), pages 204–213, Taipei, Taiwan. Asian Federation of Natural Language Processing.
Cite (Informal):
Distributional Modeling on a Diet: One-shot Word Learning from Text Only (Wang et al., IJCNLP 2017)
Copy Citation:
PDF:
https://aclanthology.org/I17-1021.pdf