Adapter Pruning using Tropical Characterization

Rishabh Bhardwaj, Tushar Vaidya, Soujanya Poria


Abstract
Adapters are widely popular parameter-efficient transfer learning approaches in natural language processing that insert trainable modules in between layers of a pre-trained language model. Apart from several heuristics, however, there has been a lack of studies analyzing the optimal number of adapter parameters needed for downstream applications. Thus, we propose an adapter pruning approach by studying the tropical characteristics of trainable modules. We cast it as an optimization problem that aims to prune parameters from the adapter layers without changing the orientation of underlying tropical hypersurfaces. Our experiments on five NLP datasets show that tropical geometry tends to identify more relevant parameters to prune when compared with the magnitude-based baseline, while a combined approach works best across the tasks.
Anthology ID:
2023.findings-emnlp.116
Volume:
Findings of the Association for Computational Linguistics: EMNLP 2023
Month:
December
Year:
2023
Address:
Singapore
Editors:
Houda Bouamor, Juan Pino, Kalika Bali
Venue:
Findings
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
1699–1706
Language:
URL:
https://aclanthology.org/2023.findings-emnlp.116
DOI:
10.18653/v1/2023.findings-emnlp.116
Bibkey:
Cite (ACL):
Rishabh Bhardwaj, Tushar Vaidya, and Soujanya Poria. 2023. Adapter Pruning using Tropical Characterization. In Findings of the Association for Computational Linguistics: EMNLP 2023, pages 1699–1706, Singapore. Association for Computational Linguistics.
Cite (Informal):
Adapter Pruning using Tropical Characterization (Bhardwaj et al., Findings 2023)
Copy Citation:
PDF:
https://aclanthology.org/2023.findings-emnlp.116.pdf