Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

Reusing Pretrained Models by Multi-linear Operators for Efficient Training

About

Training large models from scratch usually costs a substantial amount of resources. Towards this problem, recent studies such as bert2BERT and LiGO have reused small pretrained models to initialize a large model (termed the ``target model''), leading to a considerable acceleration in training. Despite the successes of these previous studies, they grew pretrained models by mapping partial weights only, ignoring potential correlations across the entire model. As we show in this paper, there are inter- and intra-interactions among the weights of both the pretrained and the target models. As a result, the partial mapping may not capture the complete information and lead to inadequate growth. In this paper, we propose a method that linearly correlates each weight of the target model to all the weights of the pretrained model to further enhance acceleration ability. We utilize multi-linear operators to reduce computational and spacial complexity, enabling acceptable resource requirements. Experiments demonstrate that our method can save 76\% computational costs on DeiT-base transferred from DeiT-small, which outperforms bert2BERT by +12.0\% and LiGO by +20.7\%, respectively.

Yu Pan, Ye Yuan, Yichun Yin, Zenglin Xu, Lifeng Shang, Xin Jiang, Qun Liu• 2023

Related benchmarks

TaskDatasetResultRank
Natural Language UnderstandingGLUE
SST-292.71
452
Image ClassificationCIFAR100
Accuracy90.23
331
ClassificationCars
Accuracy91.83
314
Image ClassificationFlowers
Accuracy97.49
127
Image ClassificationCIFAR10
Accuracy99.13
125
Image ClassificationChestXRay8
Accuracy0.5546
9
Question AnsweringSQuAD v1.1 & v2.0
SQuAD v1.1 F190.17
5
Showing 7 of 7 rows

Other info

Follow for update