|
所在平台: Udemy |
课程主页: https://www.udemy.com/course/explainable-ai-xai-for-generative-ai/
课程评论：没有评论
课程名称：可解释性人工智能（XAI）与生成性人工智能 课程概述： “可解释性人工智能（XAI）与生成性人工智能”是一门全面的课程，旨在揭示生成性人工智能的“黑箱”。该课程适合数据科学家、机器学习工程师和人工智能爱好者，帮助学员构建和部署透明、负责任及可信赖的生成性AI解决方案。 课程内容包括： - 学习生成性AI框架的现状，了解其与传统大型语言模型（LLMs）的区别以及应用场景。 - 通过Hugging Face平台，掌握访问和使用各种预训练模型的实践经验。 - 掌握基本的提示工程技术，以有效引导生成性模型产生可预测的结果。 - 深入了解可解释性人工智能（XAI）的基本概念、重要性及在文本和图像生成中的独特挑战。 - 学习在文本和条件生成系统中实施XAI的实用方法，包括注意力可视化、潜在空间分析及后期可解释性工具（如LIME和SHAP）。 - 探索如何通过提示工程实现XAI，设计能够引导模型输出并促使透明推理的提示策略，例如链式推理等。 课程结束时，学员将掌握构建更具可解释性、责任感及人性化的生成性AI系统的技能，准备在生产环境和高风险应用中使用。 授课教师背景： 教师拥有牛津大学的地理与环境硕士学位，以及剑桥大学的热带生态与保护数据科学博士学位，拥有多年的真实数据分析经验，并在国际同行评审期刊上发表过多篇论文。
Course Description:Unlock the black box of Generative AI with "Explainable AI (XAI) for Generative AI", a comprehensive course designed to bridge the gap between cutting-edge generative models and responsible, interpretable AI systems. Whether you're a data scientist, ML engineer, or AI enthusiast, this course will empower you to build and deploy transparent, accountable, and trustworthy GenAI solutions.You'll begin by exploring the landscape of Generative AI frameworks, understanding how they differ from traditional Large Language Models (LLMs), and when to use each. You'll get hands-on experience with Hugging Face, the leading open-source platform for accessing and working with pre-trained models across a wide variety of generative tasks.The course introduces basic prompt engineering techniques to guide generative models effectively and predictably. From there, you'll dive into the fundamentals of Explainable AI (XAI)-what it is, why it matters, and the unique challenges it presents in generative contexts like text and image generation.You'll learn practical methods for implementing XAI in both text-based and conditional generative systems, including techniques like attention visualization, latent space analysis, and post-hoc explainability tools such as LIME and SHAP. Finally, you'll discover how to operationalize XAI through prompt engineering, crafting prompts that not only guide model output but also elicit transparent reasoning via Chain of Thought and other explainability-oriented prompting strategies.By the end of the course, you'll have the skills to build more interpretable, responsible, and human-aligned generative AI systems-ready for use in production environments and high-stakes applications.Why Should You Take My Course?I have an MPhil (Geography and Environment) from the University of Oxford, UK. I also completed a data science PhD (Tropical Ecology and Conservation) at Cambridge University.I have several years of experience analyzing real-life data from different sources and producing publications for international peer-reviewed journals.
