Group and Shuffle: Efficient Structured Orthogonal Parametrization
Abstract
The increasing size of neural networks has led to a growing demand for methods of efficient fine-tuning. Recently, an orthogonal fine-tuning paradigm was introduced that uses orthogonal matrices for adapting the weights of a pretrained model. In this paper, we introduce a new class of structured matrices, which unifies and generalizes structured classes from previous works. We examine properties of this class and build a structured orthogonal parametrization upon it. We then use this parametrization to modify the orthogonal fine-tuning framework, improving parameter and computational efficiency. We empirically validate our method on different domains, including adapting of text-to-image diffusion models and downstream task fine-tuning in language modeling. Additionally, we adapt our construction for orthogonal convolutions and conduct experiments with 1-Lipschitz neural networks.
Cite
@article{arxiv.2406.10019,
title = {Group and Shuffle: Efficient Structured Orthogonal Parametrization},
author = {Mikhail Gorbunov and Nikolay Yudin and Vera Soboleva and Aibek Alanov and Alexey Naumov and Maxim Rakhuba},
journal= {arXiv preprint arXiv:2406.10019},
year = {2024}
}