English

AUBER: Automated BERT Regularization

Artificial Intelligence 2021-09-15 v1

Abstract

How can we effectively regularize BERT? Although BERT proves its effectiveness in various downstream natural language processing tasks, it often overfits when there are only a small number of training instances. A promising direction to regularize BERT is based on pruning its attention heads based on a proxy score for head importance. However, heuristic-based methods are usually suboptimal since they predetermine the order by which attention heads are pruned. In order to overcome such a limitation, we propose AUBER, an effective regularization method that leverages reinforcement learning to automatically prune attention heads from BERT. Instead of depending on heuristics or rule-based policies, AUBER learns a pruning policy that determines which attention heads should or should not be pruned for regularization. Experimental results show that AUBER outperforms existing pruning methods by achieving up to 10% better accuracy. In addition, our ablation study empirically demonstrates the effectiveness of our design choices for AUBER.

Keywords

Cite

@article{arxiv.2009.14409,
  title  = {AUBER: Automated BERT Regularization},
  author = {Hyun Dong Lee and Seongmin Lee and U Kang},
  journal= {arXiv preprint arXiv:2009.14409},
  year   = {2021}
}
R2 v1 2026-06-23T18:53:55.241Z