Semantic Classification and Learning Using a Linear Tranformation Model in a Probabilistic Type Theory with Records

Staffan Larsson, Jean-Philippe Bernardy


Abstract
Starting from an existing account of semantic classification and learning from interaction formulated in a Probabilistic Type Theory with Records, encompassing Bayesian inference and learning with a frequentist flavour, we observe some problems with this account and provide an alternative account of classification learning that addresses the observed problems. The proposed account is also broadly Bayesian in nature but instead uses a linear transformation model for classification and learning.
Anthology ID:
2021.reinact-1.3
Volume:
Proceedings of the Reasoning and Interaction Conference (ReInAct 2021)
Month:
October
Year:
2021
Address:
Gothenburg, Sweden
Editors:
Christine Howes, Simon Dobnik, Ellen Breitholtz, Stergios Chatzikyriakidis
Venue:
ReInAct
SIG:
SIGSEM
Publisher:
Association for Computational Linguistics
Note:
Pages:
14–22
Language:
URL:
https://aclanthology.org/2021.reinact-1.3
DOI:
Bibkey:
Cite (ACL):
Staffan Larsson and Jean-Philippe Bernardy. 2021. Semantic Classification and Learning Using a Linear Tranformation Model in a Probabilistic Type Theory with Records. In Proceedings of the Reasoning and Interaction Conference (ReInAct 2021), pages 14–22, Gothenburg, Sweden. Association for Computational Linguistics.
Cite (Informal):
Semantic Classification and Learning Using a Linear Tranformation Model in a Probabilistic Type Theory with Records (Larsson & Bernardy, ReInAct 2021)
Copy Citation:
PDF:
https://aclanthology.org/2021.reinact-1.3.pdf