課程名稱Course Title (中文) 生成式人工智慧 (英文) 開課單位Departments 資訊工程研究所 課程代碼Course No. E5790 授課教師Instructor 許超雲 學分數Credit 3.0 必/選修core required/optional 選修 開課年級Level 研究所 先修科目或先備能力(Course Pre-requisites)：神經網路概論 或 人工神經系統 課程概述與目標(Course Overview and Goals)：人工智慧今年發展的非常迅速,已經由圖形識別，轉向生成式。本課程目前主要的核心技術就是生成式人工智慧；包含了三大技術，其一為：Variational Auto encoder(VAE);其二為GAN(Generative Adversarial Networks);其三則為 denoising diffusion model. 外加 transformer 及大型語言模式等. 目標就是讓同學透過本課程對於新的生成式人工技術的技術能夠有深入了解，也可具有應用的能力 教科書(Textbook) 無 參考教材(Reference) IEEE 等論文
 課程大綱 Syllabus 學生學習目標Learning Objectives 單元學習活動Learning Activities 學習成效評量Evaluation 備註Notes 序No. 單元主題Unit topic 內容綱要Content summary 1 Generative A I Introduction to Genrative AI What is Generative AI? What is the difference between tranditional AI　and Generative AI? 討論講授閱讀討論 作業 Basic understanding 2 Variational Autoencoder (VAE) and Laternal Space (I) The review of VAE and Laternal Space What is Laternal space and its intrinsities 討論講授閱讀討論 3 From VAE To U-NET introduction of U-NET What is U-Net? 講授閱讀討論 作業 U net 4 Diffusion Model 1. Introduction of Diffusion model 2. Denoising Diffusion Probabilistic Models (DDPMs) What is diffusion model? What is it's advantages? 講授閱讀討論 作業 5 Diffusion Model (II) 3.Score-Based Generative Models (SGMs) 4. Stochastic Differential Equations (Score SDEs) What is SGMs and Score SDEs? 講授心得發表閱讀討論 6 Diffusion Model (III) stable diffusion and its applications explore stable diffussion and potential applications 講授閱讀討論 Bark understanding 7 Recent Development on Generative Adversarial Network Recent Development on Generative Adversarial Network Recent Development on Generative Adversarial Network 講授閱讀討論 8 Minterm Minterm Minterm 期中考 9 Vision Transformer CNN+ Self Attension what is self attension? 講授閱讀討論 10 Large Language model (LLM) 4 basis components of Transformer-based LLM: Word Embedding, Positional Encoding, Transformers, and Text Generation. 4 basis components of Transformer-based LLM: Word Embedding, Positional Encoding, Transformers, and Text Generation. 講授閱讀討論 11 Large Language model (LLM) (II) Retrieval - Augumented Generation (RAG ) What is RAG　and its advantages? 演講閱讀討論 作業 12 Open sources LLM Open source LLM Open Source LLM 講授閱讀討論 13 Recent Applications of LLM Recent Applications of LLM Recent Applications of LLM 講授閱讀討論 14 Explainable AI: Explainable AI: Visualizing Attention in Transformers 講授閱讀討論 15 Case Studies Aplications of GAI Aplications of GAI 講授閱讀討論 16 Final exam Final Exam Final Exam 期末考 17 彈性教學 Applications of GAI 請業界介紹Applications of GAI Applications of GAI 演講 彈性教學 18 彈性教學 Applications of GAI(II) 業界介紹Applications of GAI Applications of GAI 演講 彈性教學 心得報告

 序No. 實施期間Period 實施方式Content 教學說明Teaching instructions 彈性教學評量方式Evaluation 備註Notes 1 起:2024-12-30 迄:2025-01-12 邀請業界專家分享GAI 應用 心得報告

 教學要點概述： 1.自編教材 Handout by Instructor： ■ 1-1.簡報 Slids □ 1-2.影音教材 Videos □ 1-3.教具 Teaching Aids □ 1-4.教科書 Textbook □ 1-5.其他 Other □ 2.自編評量工具/量表 Educational Assessment □ 3.教科書作者提供 Textbook 成績考核 Performance Evaluation： 期末考：30%   期中考：30%   彈性教學：10%   作業：30%   教學資源(Teaching Resources)： ■ 教材電子檔(Soft Copy of the Handout or the Textbook) □ 課程網站(Website) 扣考規定：https://curri.ttu.edu.tw/p/412-1033-1254.php