Summary of Domain Adaptation For Efficiently Fine-tuning Vision Transformer with Encrypted Images, by Teru Nagamori et al.
Domain Adaptation for Efficiently Fine-tuning Vision Transformer with Encrypted Images
by Teru Nagamori, Sayaka Shiota, Hitoshi Kiya
First submitted to arxiv on: 5 Sep 2023
Categories
- Main: Computer Vision and Pattern Recognition (cs.CV)
- Secondary: Cryptography and Security (cs.CR); Machine Learning (cs.LG)
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 research proposes a novel approach to fine-tuning deep neural networks (DNNs) for privacy-preserving learning, access control, and adversarial defenses. The method utilizes the vision transformer (ViT) and does not degrade model accuracy when working with transformed images. The proposed domain adaptation method is based on the embedding structure of ViT and demonstrates effectiveness in preventing accuracy degradation using encrypted images with CIFAR-10 and CIFAR-100 datasets. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper helps us learn how to make computers smarter without putting our personal information at risk. By fine-tuning a special kind of computer model called a vision transformer, we can keep our data private while still getting good results. This is important because it lets us use computers for things like security and defense without worrying about our secrets being revealed. |
Keywords
* Artificial intelligence * Domain adaptation * Embedding * Fine tuning * Vision transformer * Vit
