Revisiting the Hierarchical Multiscale LSTM

Ákos Kádár, Marc-Alexandre Côté, Grzegorz Chrupała, Afra Alishahi


Abstract
Hierarchical Multiscale LSTM (Chung et. al., 2016) is a state-of-the-art language model that learns interpretable structure from character-level input. Such models can provide fertile ground for (cognitive) computational linguistics studies. However, the high complexity of the architecture, training and implementations might hinder its applicability. We provide a detailed reproduction and ablation study of the architecture, shedding light on some of the potential caveats of re-purposing complex deep-learning architectures. We further show that simplifying certain aspects of the architecture can in fact improve its performance. We also investigate the linguistic units (segments) learned by various levels of the model, and argue that their quality does not correlate with the overall performance of the model on language modeling.
Anthology ID:
C18-1272
Volume:
Proceedings of the 27th International Conference on Computational Linguistics
Month:
August
Year:
2018
Address:
Santa Fe, New Mexico, USA
Editors:
Emily M. Bender, Leon Derczynski, Pierre Isabelle
Venue:
COLING
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
3215–3227
Language:
URL:
https://aclanthology.org/C18-1272
DOI:
Bibkey:
Cite (ACL):
Ákos Kádár, Marc-Alexandre Côté, Grzegorz Chrupała, and Afra Alishahi. 2018. Revisiting the Hierarchical Multiscale LSTM. In Proceedings of the 27th International Conference on Computational Linguistics, pages 3215–3227, Santa Fe, New Mexico, USA. Association for Computational Linguistics.
Cite (Informal):
Revisiting the Hierarchical Multiscale LSTM (Kádár et al., COLING 2018)
Copy Citation:
PDF:
https://preview.aclanthology.org/emnlp22-frontmatter/C18-1272.pdf
Data
Penn Treebank