Federated Learning （FL） is a new distributed machine learning paradigm that breaks down data barriers and protects data privacy at the same time， thereby enabling clients to collaboratively train a machine learning model without sharing local data. However， how to deal with Non-Independent Identical Distribution （Non-IID） data from different clients remains a huge challenge faced by FL. Some existing proposed solutions to this problem do not utilize the implicit relationship between local and global models to solve the problem simply and efficiently. To address the Non-IID issue of different clients in FL， novel FL optimization algorithms including Federated Self-Regularization （FedSR） and Dynamic Federated Self-Regularization （Dyn-FedSR） were proposed. In FedSR， self-regularization penalty terms were introduced in each training round to modify the local loss function dynamically， and by building a relationship between the local and the global models， the local model was closer to the global model that aggregates rich knowledge， thereby alleviating the client drift problem caused by Non-IID data. In Dyn-FedSR， the self-regularization term coefficient was determined dynamically by calculating the similarity between the local and global models. Extensive experimental analyses on different tasks demonstrate that the two algorithms， FedSR and Dyn-FedSR， significantly outperform the state-of-the-art FL algorithms such as Federated Averaging （FedAvg） algorithm， Federated Proximal （FedProx） optimization algorithm and Stochastic Controlled Averaging algorithm （SCAFFOLD） in various scenarios， and can achieve efficient communication and high accuracy， as well as the robustness to imbalanced data and uncertain local updates.

Most of the existing methods combining Generative Adversarial Network （GAN） and differential privacy use gradient perturbation to achieve privacy protection， that is in the process of optimization， the gradient clipping technology was used to constrain the sensitivity of the optimizer to single data， and random noise is added to the clipped gradient to achieve the purpose of model protection. However， most methods take the clipping threshold as a fixed parameter during training. Whether the threshold is too large or too small， the performance of the model will be affected. To solve this problem， DGC_DPGAN （Dynamic Gradient Clipping Differential Privacy Generative Adversarial Network） with dynamic gradient threshold clipping was proposed to consider privacy protection and model performance at the same time. In this algorithm， combined with the pre-training technology， in the process of optimization， the mean gradient F-norm value of each batch of privacy data was obtained as the dynamic gradient clipping threshold at first， and then the gradient was perturbed. Considering different clipping orders， CLIP_DGC_DPGAN （Clip Dynamic Gradient Clipping Differential Privacy Generative Adversarial Network）， which clipping first and adding noise after， and DGC_DPGAN， which adding noise first and clipping after， were proposed， and Rényi Accountant was used to calculate the privacy loss. Experimental results show that under the same privacy budget， the two proposed dynamic gradient clipping algorithms are better than the fixed gradient threshold clipping method. On Mnist dataset， the two proposed algorithm has the Inception Score （IS）， Structural SIMilarity （SSIM）， and Convolutional Neural Network （CNN） classification accuracy improved by 0.32 to 3.92， 0.03 to 0.27， and 7% to 44% respectively； on Fashion-Mnist dataset， the two proposed algorithm has the IS， SSIM， and CNN classification accuracy improved by 0.40 to 4.32， 0.01 to 0.44 and 20% to 51% respectively. At the same time， the usability of the images generated by GAN model is better.