0

The Expressive Power of Low-Rank Adaptation

Low-Rank Adaptation (LoRA)理论上被分析,证明其在特定条件下能有效调整模型以匹配小型目标模型,并量化了其近似误差。

Year
2023
Venue
arXiv 2023
Authors
2
Hosting
Abstract onlyARXIV-DEFAULT

Cite

Notes

Only stored in your browser.

Attribution

Abstract & full text
arxiv.org/abs/2310.17513v3ARXIV-DEFAULT
TL;DR
Semantic Scholar
Attribution policy →

Abstract

Low-Rank Adaptation (LoRA), a parameter-efficient fine-tuning method that leverages low-rank adaptation of weight matrices, has emerged as a prevalent technique for fine-tuning pre-trained models such as large language models and diffusion models. Despite its huge success in practice, the theoretical underpinnings of LoRA have largely remained unexplored. This paper takes the first step to bridge this gap by theoretically analyzing the expressive power of LoRA. We prove that, for fully connected neural networks, LoRA can adapt any model $f$ to accurately represent any smaller target model $\overline{f}$ if LoRA-rank $\geq(\text{width of }f) \times \frac{\text{depth of }\overline{f}}{\text{depth of }f}$. We also quantify the approximation error when LoRA-rank is lower than the threshold. For Transformer networks, we show any model can be adapted to a target model of the same size with rank-$(\frac{\text{embedding size}}{2})$ LoRA adapters.

Authors

2