|
所在平台: Udemy |
课程主页: https://www.udemy.com/course/federated-learning-theory-and-practical/
课程评论：没有评论
课程名称：联邦学习：理论与实践 课程概述： “联邦学习：理论与实践”课程旨在为您提供关于机器学习中一个令人兴奋且不断发展的领域——联邦学习（FL）的全面介绍。在数据隐私日益重要的时代，FL通过使机器学习模型能够在去中心化的数据源（如智能手机或本地客户端）上进行训练，而无需共享敏感数据，提供了一种解决方案。 课程内容将从机器学习的基础知识开始，确保您具备扎实的基础。接下来，您将深入了解联邦学习的核心概念，包括其发展的动机、不同类型（水平、垂直和结合FL），以及它与传统机器学习方法的比较。到第三周，您不仅将掌握理论知识，还将准备好从零开始实现FL系统，并使用诸如FLOWER等流行框架。 此外，您还将探索先进的主题，如隐私增强技术，包括差分隐私和同态加密，并深入了解实际挑战，如客户端选择和梯度反演攻击。无论您是数据科学家、机器学习工程师，还是对隐私保护AI感兴趣的人，这门课程都将提供必要的理论基础和实践技能，帮助您应对联邦学习的新兴领域。
"Federated Learning: Theory and Practical" is designed to provide you with a comprehensive introduction to one of the most exciting and evolving areas in machine learning-federated learning (FL). In an era where data privacy is becoming increasingly important, FL offers a solution by enabling machine learning models to be trained across decentralized data sources, such as smartphones or local clients, without the need to share sensitive data.This course starts with the basics of machine learning to ensure a solid foundation. You will then dive into the core concepts of federated learning, including the motivations behind its development, the different types (horizontal, vertical, and combined FL), and how it compares to traditional machine learning approaches.By week three, you'll not only grasp the theory but also be ready to implement FL systems from scratch and using popular frameworks like FLOWER. You'll explore advanced topics such as privacy-enhancing technologies, including differential privacy and homomorphic encryption, and gain insight into practical challenges like client selection and gradient inversion attacks.Whether you are a data scientist, machine learning engineer, or someone curious about privacy-preserving AI, this course offers the theoretical grounding and hands-on skills necessary to navigate the emerging landscape of federated learning.
