Sparse Progressive Distillation: Resolving Overfitting under Pretrain-and-Finetune Paradigm

Huang, Shaoyi, Xu, Dongkuan, Yen, Ian E. H., Wang, Yijue, Chang, Sung-en, Li, Bingbing, Chen, Shiyang, Xie, Mimi, Rajasekaran, Sanguthevar, Liu, Hang, Ding, Caiwen

arXiv.org Artificial Intelligence 

Conventional wisdom in pruning Transformer-based language models is that pruning reduces the model expressiveness and thus is more likely to underfit rather than overfit. However, under the trending pretrain-and-finetune paradigm, we postulate a counter-traditional hypothesis, that is: pruning increases the risk of overfitting when performed at the fine-tuning phase. In this paper, we aim to address the overfitting problem and improve pruning performance via progressive knowledge distillation with error-bound properties. We show for the first time that reducing the risk of overfitting can help the effectiveness of pruning under the pretrain-and-finetune paradigm. Ablation studies and experiments on the GLUE benchmark show that our method outperforms the leading competitors across different tasks.

Duplicate Docs Excel Report

Title
None found

Similar Docs  Excel Report  more

TitleSimilaritySource
None found