Saved in:
Bibliographic Details
Main Authors: Nguyen, John, Wang, Sid, Li, Ke, Wu, Carole-Jean
Format: Preprint
Published: 2023
Subjects:
Online Access:https://arxiv.org/abs/2305.15348
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914963710279680
author Nguyen, John
Wang, Sid
Li, Ke
Wu, Carole-Jean
author_facet Nguyen, John
Wang, Sid
Li, Ke
Wu, Carole-Jean
contents Fine-tuning large-scale Transformers has led to the explosion of many AI applications across Natural Language Processing and Computer Vision tasks. However, fine-tuning all pre-trained model parameters becomes impractical as the model size and number of tasks increase. Parameter-efficient transfer learning (PETL) methods aim to address these challenges. While effective in reducing the number of trainable parameters, PETL methods still require significant energy and computational resources to fine-tune. In this paper, we introduce \textbf{RE}current \textbf{AD}aption (READ) -- a lightweight and memory-efficient fine-tuning method -- to overcome the limitations of the current PETL approaches. Specifically, READ inserts a small RNN network alongside the backbone model so that the model does not have to back-propagate through the large backbone network. Through comprehensive empirical evaluation of the GLUE benchmark, we demonstrate READ can achieve a $56\%$ reduction in the training memory consumption and an $84\%$ reduction in the GPU energy usage while retraining high model quality compared to full-tuning. Additionally, the model size of READ does not grow with the backbone model size, making it a highly scalable solution for fine-tuning large Transformers.
format Preprint
id arxiv_https___arxiv_org_abs_2305_15348
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle READ: Recurrent Adaptation of Large Transformers
Nguyen, John
Wang, Sid
Li, Ke
Wu, Carole-Jean
Machine Learning
Artificial Intelligence
Fine-tuning large-scale Transformers has led to the explosion of many AI applications across Natural Language Processing and Computer Vision tasks. However, fine-tuning all pre-trained model parameters becomes impractical as the model size and number of tasks increase. Parameter-efficient transfer learning (PETL) methods aim to address these challenges. While effective in reducing the number of trainable parameters, PETL methods still require significant energy and computational resources to fine-tune. In this paper, we introduce \textbf{RE}current \textbf{AD}aption (READ) -- a lightweight and memory-efficient fine-tuning method -- to overcome the limitations of the current PETL approaches. Specifically, READ inserts a small RNN network alongside the backbone model so that the model does not have to back-propagate through the large backbone network. Through comprehensive empirical evaluation of the GLUE benchmark, we demonstrate READ can achieve a $56\%$ reduction in the training memory consumption and an $84\%$ reduction in the GPU energy usage while retraining high model quality compared to full-tuning. Additionally, the model size of READ does not grow with the backbone model size, making it a highly scalable solution for fine-tuning large Transformers.
title READ: Recurrent Adaptation of Large Transformers
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2305.15348