Summary of Parameter-efficient Fine-tuning with Discrete Fourier Transform, by Ziqi Gao et al.
Parameter-Efficient Fine-Tuning with Discrete Fourier Transform
by Ziqi Gao, Qichao Wang, Aochuan Chen, Zijing Liu, Bingzhe Wu, Liang Chen, Jia Li
First submitted to arxiv on: 5 May 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary This paper introduces FourierFT, a method that compresses trainable parameters in fine-tuning foundation models using the Fourier transform. By treating the weight change as a matrix and learning only a small fraction of its spectral coefficients, FourierFT reduces the number of trainable parameters while maintaining performance. The authors demonstrate comparable or better results than LoRA on various tasks, including natural language understanding, generation, instruction tuning, and image classification, with fewer parameters. For example, when fine-tuning the LLaMA2-7B model for instruction tuning, FourierFT outperforms LoRA with only 0.064M trainable parameters compared to LoRA’s 33.5M. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper makes a big discovery that helps computers learn new skills without using too much memory. It’s like a special trick to make the computer understand things better. The researchers created a new method called FourierFT, which uses something called the Fourier transform to shrink the amount of information needed to learn new tasks. This means that computers can do more tasks with less memory, making them faster and more efficient. |
Keywords
» Artificial intelligence » Fine tuning » Image classification » Instruction tuning » Language understanding » Lora
