Critical Learning Periods: Leveraging Early Training Dynamics for Efficient Data Pruning
Everlyn Chimoto, Jay Gala, Orevaoghene Ahia, Julia Kreutzer, Bruce Bassett, Sara Hooker
Abstract
Neural Machine Translation models are extremely data and compute-hungry. However, not all datapoints contribute equally to model training and generalization. Data pruning to remove the low-value data points has the benefit of drastically reducing the compute budget without significantdrop in model performance. In this paper, we propose a new data pruning technique: CheckpointsAcross Time (CAT ), that leverages early model training dynamics to identify the most relevantdata points for model performance. We benchmark CAT against several data pruning techniquesincluding COMET-QE, LASER and LaBSE. We find that CAT outperforms the benchmarks onIndo-European languages on multiple test sets. When applied to English-German, English-Frenchand English-Swahili translation tasks, CAT achieves comparable performance to using the fulldataset, while pruning up to 50% of training data. We inspect the data points that CAT selectsand find that it tends to favour longer sentences and sentences with unique or rare words.- Anthology ID:
- 2024.findings-acl.560
- Volume:
- Findings of the Association for Computational Linguistics: ACL 2024
- Month:
- August
- Year:
- 2024
- Address:
- Bangkok, Thailand
- Editors:
- Lun-Wei Ku, Andre Martins, Vivek Srikumar
- Venue:
- Findings
- SIG:
- Publisher:
- Association for Computational Linguistics
- Note:
- Pages:
- 9407–9426
- Language:
- URL:
- https://aclanthology.org/2024.findings-acl.560
- DOI:
- 10.18653/v1/2024.findings-acl.560
- Cite (ACL):
- Everlyn Chimoto, Jay Gala, Orevaoghene Ahia, Julia Kreutzer, Bruce Bassett, and Sara Hooker. 2024. Critical Learning Periods: Leveraging Early Training Dynamics for Efficient Data Pruning. In Findings of the Association for Computational Linguistics: ACL 2024, pages 9407–9426, Bangkok, Thailand. Association for Computational Linguistics.
- Cite (Informal):
- Critical Learning Periods: Leveraging Early Training Dynamics for Efficient Data Pruning (Chimoto et al., Findings 2024)
- PDF:
- https://preview.aclanthology.org/dois-2013-emnlp/2024.findings-acl.560.pdf