UniDrop: A Simple yet Effective Technique to Improve Transformer without Extra Cost
arXiv:2104.04946
Abstract
Transformer architecture achieves great success in abundant natural language processing tasks. The over-parameterization of the Transformer model has motivated plenty of works to alleviate its overfitting for superior performances. With some explorations, we find simple techniques such as dropout, can greatly boost model performance with a careful design. Therefore, in this paper, we integrate different dropout techniques into the training of Transformer models. Specifically, we propose an approach named UniDrop to unites three different dropout techniques from fine-grain to coarse-grain, i.e., feature dropout, structure dropout, and data dropout. Theoretically, we demonstrate that these three dropouts play different roles from regularization perspectives. Empirically, we conduct experiments on both neural machine translation and text classification benchmark datasets. Extensive results indicate that Transformer with UniDrop can achieve around 1.5 BLEU improvement on IWSLT14 translation tasks, and better accuracy for the classification even using strong pre-trained RoBERTa as backbone.
arXiv commentAccepted by NAACL 2021
References in corpus (9)
- DropBlock: A regularization method for convolutional networks
- Pay Less Attention with Lightweight and Dynamic Convolutions
- Incorporating BERT into Neural Machine Translation
- Zoneout: Regularizing RNNs by Randomly Preserving Hidden Activations
- Weighted Transformer Network for Machine Translation
- Understanding and Improving Transformer From a Multi-Particle Dynamic System Point of View
- Improving Neural Language Modeling via Adversarial Training
- Robust Learning with Jacobian Regularization
- The Implicit and Explicit Regularization Effects of Dropout