Unsupervised Federated Learning for Privacy Preserving in Face Recognition System
Keywords: Unsupervised Federated Learning for Face Recognition in Decentralized Environments
Abstract: Recent advancements in face recognition involve training on a single computer, often containing sensitive personal information, raising privacy concerns. To address this, attention turns to federated learning for unsupervised face recognition, leveraging decentralized edge devices. Each device independently undergoes model training, transmitting results to a secure aggregator. We utilize GANs to diversify data without the need for transmission, thereby preserving privacy throughout the entire process. The aggregator integrates these diverse models into a single global model, which is then transmitted back to the edge devices for continued improvement. Experiments on CelebA datasets demonstrate that federated learning not only preserves privacy but also maintains high levels of performance.
Primary Area: unsupervised, self-supervised, semi-supervised, and supervised representation learning
Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics.
Submission Guidelines: I certify that this submission complies with the submission instructions as described on https://iclr.cc/Conferences/2025/AuthorGuide.
Reciprocal Reviewing: I understand the reciprocal reviewing requirement as described on https://iclr.cc/Conferences/2025/CallForPapers. If none of the authors are registered as a reviewer, it may result in a desk rejection at the discretion of the program chairs. To request an exception, please complete this form at https://forms.gle/Huojr6VjkFxiQsUp6.
Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors’ identity.
No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.
Submission Number: 13104
Loading