# Privacy-Preserving Split Learning via Patch Shuffling over Transformers
> Patch shuffling is a privacy-preserving split learning technique for vision transformers. It randomly permutes image patches before intermediate feature sharing to reduce feature inversion risk while preserving transformer accuracy through permutation equivariance.
Type: paper
Venue: ICDM 2022
Date: 2022
Authors: Dixi Yao, Liyao Xiang, Hengyuan Xu, Hangyu Ye, Yingqi Chen
## Direct Links
- [Canonical topic page](https://dixiyao.github.io/topics/patch-shuffling/)
- [Paper](https://ieeexplore.ieee.org/abstract/document/10027647)
- [PDF](https://dixiyao.github.io/assests/papers/Privacy-Preserving-Split-Learning-via-Patch-Shuffling-over-Transformers.pdf)
- [Code](https://github.com/dixiyao/PatchShuffling)
- [Slides](https://dixiyao.github.io/assests/slides/PatchShuffle.pdf)
## Search Queries and Aliases
- patch shuffling
- PatchShuffling
- split learning privacy
- privacy-preserving split learning
- transformer patch permutation
- feature inversion defense
- vision transformer privacy
- ICDM patch shuffling
## Search Intents
- patch shuffling split learning paper
- PatchShuffling code
- privacy-preserving split learning transformers
- feature inversion defense vision transformer