FedBoosting：面向文本识别的采用梯度保护提升的联邦学习
FedBoosting: Federated Learning with Gradient Protected Boosting for Text Recognition
AI总结：
针对联邦学习在非独立同分布数据下泛化差、易梯度泄露的问题，提出结合同态加密与差分隐私的FedBoosting算法，在文本识别任务上实现了准确率和效率的双重提升。
摘要：
传统机器学习方法需要集中式数据来训练模型，但在因隐私保护、梯度保护等因素存在数据共享限制的场景下，这一要求往往无法满足。近期提出的联邦学习（Federated Learning, FL）框架支持在数据无需集中、数据所有者之间不共享数据的前提下，协同学习一个共享模型。但本文发现，在非独立同分布（Non-Independent and Non-Identically Distributed, Non-IID）数据上，联合模型的泛化能力较差，尤其是采用联邦平均（Federated Averaging, FedAvg）策略时，会因权重发散现象加剧这一问题。为此，我们提出一种适用于联邦学习的新型提升算法，既能解决泛化能力不足和梯度泄露问题，又能在基于梯度的优化中实现更快收敛。此外，我们引入了一种结合同态加密（Homomorphic Encryption, HE）和差分隐私（Differential Privacy, DP）的安全梯度共享协议，可抵御梯度泄露攻击，且无需使用扩展性差的成对加密方式。我们在公开基准的视觉文本识别任务上验证了所提的Federated Boosting（FedBoosting）方法，结果表明该方法在预测准确率和运行效率上均有显著提升。
英文摘要：
Typical machine learning approaches require centralized data for model training, which may not be possible where restrictions on data sharing are in place due to, for instance, privacy and gradient protection. The recently proposed Federated Learning (FL) framework allows learning a shared model collaboratively without data being centralized or shared among data owners. However, we show in this paper that the generalization ability of the joint model is poor on Non-Independent and Non-Identically Distributed (Non-IID) data, particularly when the Federated Averaging (FedAvg) strategy is used due to the weight divergence phenomenon. Hence, we propose a novel boosting algorithm for FL to address both the generalization and gradient leakage issues, as well as achieve faster convergence in gradient-based optimization. In addition, a secure gradient sharing protocol using Homomorphic Encryption (HE) and Differential Privacy (DP) is introduced to defend against gradient leakage attack and avoid pairwise encryption that is not scalable. We demonstrate the proposed Federated Boosting (FedBoosting) method achieves noticeable improvements in both prediction accuracy and run-time efficiency in a visual text recognition task on public benchmark.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
