Decentralized Word2Vec Using Gossip Learning

Abdul Aziz Alkathiri, Lodovico Giaretta, Sarunas Girdzijauskas, Magnus Sahlgren


Abstract
Advanced NLP models require huge amounts of data from various domains to produce high-quality representations. It is useful then for a few large public and private organizations to join their corpora during training. However, factors such as legislation and user emphasis on data privacy may prevent centralized orchestration and data sharing among these organizations. Therefore, for this specific scenario, we investigate how gossip learning, a massively-parallel, data-private, decentralized protocol, compares to a shared-dataset solution. We find that the application of Word2Vec in a gossip learning framework is viable. Without any tuning, the results are comparable to a traditional centralized setting, with a loss of quality as low as 4.3%. Furthermore, the results are up to 54.8% better than independent local training.
Anthology ID:
2021.nodalida-main.40
Volume:
Proceedings of the 23rd Nordic Conference on Computational Linguistics (NoDaLiDa)
Month:
May 31--2 June
Year:
2021
Address:
Reykjavik, Iceland (Online)
Editors:
Simon Dobnik, Lilja Øvrelid
Venue:
NoDaLiDa
SIG:
Publisher:
Linköping University Electronic Press, Sweden
Note:
Pages:
373–377
Language:
URL:
https://aclanthology.org/2021.nodalida-main.40
DOI:
Bibkey:
Cite (ACL):
Abdul Aziz Alkathiri, Lodovico Giaretta, Sarunas Girdzijauskas, and Magnus Sahlgren. 2021. Decentralized Word2Vec Using Gossip Learning. In Proceedings of the 23rd Nordic Conference on Computational Linguistics (NoDaLiDa), pages 373–377, Reykjavik, Iceland (Online). Linköping University Electronic Press, Sweden.
Cite (Informal):
Decentralized Word2Vec Using Gossip Learning (Alkathiri et al., NoDaLiDa 2021)
Copy Citation:
PDF:
https://aclanthology.org/2021.nodalida-main.40.pdf