标题： 基于低秩适应的视觉变压器在隐私保护图像分类中的有效微调 显示英文标题

标题： Effective Fine-Tuning of Vision Transformers with Low-Rank Adaptation for Privacy-Preserving Image Classification

摘要： 我们提出了一种低秩适应方法，用于训练保护隐私的视觉变换器（ViT）模型，该方法有效地冻结了预训练ViT模型的权重。 在所提出的方法中，可训练的秩分解矩阵被注入到ViT架构的每一层中，并且与传统低秩适应方法不同，补丁嵌入层并未被冻结。 所提出的方法不仅允许减少可训练参数的数量，还能保持与全量微调几乎相同的准确性。 显示英文摘要

摘要： We propose a low-rank adaptation method for training privacy-preserving vision transformer (ViT) models that efficiently freezes pre-trained ViT model weights. In the proposed method, trainable rank decomposition matrices are injected into each layer of the ViT architecture, and moreover, the patch embedding layer is not frozen, unlike in the case of the conventional low-rank adaptation methods. The proposed method allows us not only to reduce the number of trainable parameters but to also maintain almost the same accuracy as that of full-time tuning.