@article{Gao_Xu_Shi_Ren_Yu_Liang_Jiang_Li_2022, title={AutoBERT-Zero: Evolving BERT Backbone from Scratch}, volume={36}, url={https://ojs.aaai.org/index.php/AAAI/article/view/21311}, DOI={10.1609/aaai.v36i10.21311}, abstractNote={Transformer-based pre-trained language models like BERT and its variants have recently achieved promising performance in various natural language processing (NLP) tasks. However, the conventional paradigm constructs the backbone by purely stacking the manually designed global self-attention layers, introducing inductive bias and thus leads to sub-optimal. In this work, we make the first attempt to automatically discover novel pre-trained language model (PLM) backbone on a flexible search space containing the most fundamental operations from scratch. Specifically, we propose a well-designed search space which (i) contains primitive math operations in the intra-layer level to explore novel attention structures, and (ii) leverages convolution blocks to be the supplementary for attentions in the inter-layer level to better learn local dependency. To enhance the efficiency for finding promising architectures, we propose an Operation-Priority Neural Architecture Search (OP-NAS) algorithm, which optimizes both the search algorithm and evaluation of candidate models. Specifically, we propose Operation-Priority (OP) evolution strategy to facilitate model search via balancing exploration and exploitation. Furthermore, we design a Bi-branch Weight-Sharing (BIWS) training strategy for fast model evaluation. Extensive experiments show that the searched architecture (named AutoBERT-Zero) significantly outperforms BERT and its variants of different model capacities in various downstream tasks, proving the architecture’s transfer and scaling abilities. Remarkably, AutoBERT-Zero-base outperforms RoBERTa-base (using much more data) and BERT-large (with much larger model size) by 2.4 and 1.4 higher score on GLUE test set.}, number={10}, journal={Proceedings of the AAAI Conference on Artificial Intelligence}, author={Gao, Jiahui and Xu, Hang and Shi, Han and Ren, Xiaozhe and Yu, Philip L. H. and Liang, Xiaodan and Jiang, Xin and Li, Zhenguo}, year={2022}, month={Jun.}, pages={10663-10671} }