Due to the excessive cost of large-scale language model pre-training,
considerable efforts have been made to train BERT progressively -- start from
an inferior but low-cost model and gradually grow the model to increase the
computational complexity. Our objective is to advance the understanding of
Transformer growth and discover principles that guide progressive training.
First, we find that similar to network architecture search, Transformer growth
also favors compound scaling. Specifically, while existing methods only conduct
network growth in a single dimension, we observe that it is beneficial to use
compound growth operators and balance multiple dimensions (e.g., depth, width,
and input length of the model). Moreover, we explore alternative growth
operators in each dimension via controlled comparison to give operator
selection practical guidance. In light of our analyses, the proposed method
speeds up BERT pre-training by 73.6% and 82.2% for the base and large models
respectively, while achieving comparable performances

本文旨在通过对 Transformer 增长的探索来指导渐进式训练，发现复合缩放方法可以平衡模型的深度、宽度和输入的长度，并通过比较实验探索每个维度的替代增长操作，最终使 BERT 的预训练加快 73.6％（基本模型）和 82.2％（大模型），并达到相当的性能。