“Bort: Optimal Subarchitecture Extraction For BERT”, 2020-10-20 (; similar):
We extract an optimal subset of architectural parameters for the BERT architecture from et al 2018 by applying recent breakthroughs in algorithms for neural architecture search.
This optimal subset, which we refer to as Bort, is demonstrably smaller, having an effective (that is, not counting the embedding layer) size of 5.5% the original BERT-large architecture, and 16% of the net size. Bort is also able to be pretrained in 288 GPU hours, which is 1.2% of the time required to pretrain the highest-performing BERT parametric architectural variant, RoBERTa-large ( et al 2019), and about 33% of that of the world-record, in GPU hours, required to train BERT-large on the same hardware.
It is also 7.9× faster on a CPU, as well as being better performing than other compressed variants of the architecture, and some of the non-compressed variants: it obtains performance improvements of between 0.3% and 31%, absolute, with respect to BERT-large, on multiple public natural language understanding (NLU) benchmarks.