Analyzing Bayesian Crosslingual Transfer in Topic Models

Shudong Hao, Michael J. Paul


Abstract
We introduce a theoretical analysis of crosslingual transfer in probabilistic topic models. By formulating posterior inference through Gibbs sampling as a process of language transfer, we propose a new measure that quantifies the loss of knowledge across languages during this process. This measure enables us to derive a PAC-Bayesian bound that elucidates the factors affecting model quality, both during training and in downstream applications. We provide experimental validation of the analysis on a diverse set of five languages, and discuss best practices for data collection and model design based on our analysis.
Anthology ID:
N19-1158
Volume:
Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers)
Month:
June
Year:
2019
Address:
Minneapolis, Minnesota
Venue:
NAACL
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
1551–1565
Language:
URL:
https://www.aclweb.org/anthology/N19-1158
DOI:
10.18653/v1/N19-1158
Bib Export formats:
BibTeX MODS XML EndNote
PDF:
http://aclanthology.lst.uni-saarland.de/N19-1158.pdf
Supplementary:
 N19-1158.Supplementary.pdf
Video:
 https://vimeo.com/364703295