差分隐私联邦学习从入门到发文

这是关于差分隐私联邦学习一份学习资料：有关理论知识、收敛性分析、代码、进展

一、学习相关理论

1. 差分隐私理论解读

（1）该博客包括拉普拉斯噪声、差分隐私组合定理等。
https://blog.csdn.net/MathThinker/article/details/51464273
https://blog.csdn.net/MathThinker/article/details/51637781

（2）满足差分隐私约束的深度学习系统详解
https://zhuanlan.zhihu.com/p/438563580
细读会加深你关于在深度学习中应用差分隐私的理解

2. 联邦学习相关收敛性分析

Bilibili up主丸一口
https://space.bilibili.com/3461572290677609/video?tid=36&special_type=&pn=1&keyword=&order=pubdate

3. 差分隐私经典论文解读

Bilibili up主JeffffffFu
https://www.bilibili.com/video/BV16A4y1X74k/?spm_id_from=333.999.0.0&vd_source=93b0e444c0dae60de23bb4f18190a0e3

4. 联邦学习代码解读

https://zhuanlan.zhihu.com/p/263959892?utm_source=wechat_session
该博主逐字逐句解读联邦学习模型训练、数据集分配代码

5. 深度学习相关代码网站

Papers with code
https://paperswithcode.com/task/federated-learning

二、必读论文

1.联邦学习开山之作FedAVG：Communication-Efficient Learning of Deep Networksfrom Decentralized Data
https://blog.csdn.net/qq_41605740/article/details/124584939?spm=1001.2014.3001.5501
2.将差分隐私引入深度学习：Deep Learning with Differential Privacy
https://dl.acm.org/doi/10.1145/2976749.2978318
3.瑞丽差分隐私：Rényi Differential Privacy
https://ieeexplore.ieee.org/document/8049725
4.联邦学习差分隐私收敛性分析：Federated Learning With Differential Privacy: Algorithms and Performance Analysis
https://ieeexplore.ieee.org/document/9069945

三、最新进展2023

1.联邦学习SOTA模型：Improving Generalization in Federated Learning by Seeking Flat Minima
https://dl.acm.org/doi/abs/10.1007/978-3-031-20050-2_38
2.差分隐私联邦学习SOTA模型：Make Landscape Flatter in Differentially Private Federated Learning
https://ieeexplore.ieee.org/document/10203141