Summary of Privacy-preserving Federated Learning with Consistency Via Knowledge Distillation Using Conditional Generator, by Kangyang Luo et al.
Privacy-Preserving Federated Learning with Consistency via Knowledge Distillation Using Conditional Generator
by Kangyang Luo, Shuai Wang, Xiang Li, Yunshi Lan, Ming Gao, Jinlong Shu
First submitted to arxiv on: 11 Sep 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Distributed, Parallel, and Cluster Computing (cs.DC)
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 paper proposes a novel Federated Learning (FL) method called FedMD-CG, which aims to balance high performance and high-level privacy preservation. The existing methods in FL are vulnerable to privacy leakage caused by privacy inference attacks. To address this issue, the proposed method decouples each client’s local model into a feature extractor and a classifier, using a conditional generator for server-side model aggregation. Knowledge distillation is employed to train local models and generators at both the latent feature level and the logit level. Additionally, classification losses and diversity losses are designed to enhance client-side training. The experimental results on various image classification tasks demonstrate the superiority of FedMD-CG. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper creates a new way for computers to learn together without sharing their personal data. It’s called Federated Learning (FL), and it helps keep private information safe. Some FL methods are not good at keeping things private, so this paper proposes a new method called FedMD-CG that does better. The idea is to split each computer’s learning model into two parts: one for finding features and one for classifying them. This makes it harder for attackers to figure out what’s going on. The paper also tests its method on lots of different image classification tasks and shows that it works really well. |
Keywords
» Artificial intelligence » Classification » Federated learning » Image classification » Inference » Knowledge distillation
