教學大綱與進度

課程基本資料：

學年期	課號	課程名稱	階段	學分	時數	修	教師	班級	人	撤	備註
113-2	342874	深度學習	1	3.0	3	★	林郁修	自動化所	10	1	計中電腦教室

教學大綱與進度：

教師姓名	林郁修　教師諮商時間(Office Hours)
Email	yhlin@ntut.edu.tw
最後更新時間	2025-01-17 16:32:40
課程大綱	This semester's Deep Learning (DL) course will focus on Machine Learning (ML), DL, and Generative AI in data science analytics. We will look at supervised, unsupervised and structured learning methods, which will include classic Multi-Layer Perceptron (MLP) as a shallow or Deep Neural Network (DNN), Convolutional Neural Networks (CNNs), Long Short-Term Memory (LSTM), k-Means clustering, and Generative Adversarial Networks (GANs). This course is designed for graduate students who are interested in AI and its applications in data science analytics. Students who take this course should have mathematical maturity of undergraduate curricula in science and engineering, including Calculus, Discrete Mathematics, Linear Algebra and programming language such as C/C++/MATLAB/Python. Teaching approaches conducted in this course emphasize analysis, modeling and design principles of AI methodology. Case studies, engineering applications of AI in data science analytics, will also be covered in this course. Collaboration Policy: You are free to form study groups and discuss programming assignments and projects. However, you must write up assignments and code independently, and you must acknowledge in your submission all the students you discussed with.
課程進度	(中文授課) Week Topic (PART I) 1 Introduction to IoT/AI including ML/DL/Generative AI (Overview of this course) 2 Foundations for Python programming language 3 Foundations for AI (PART II) 4 Supervised learning (I) -- Perceptron (Programming assignment #1) 5 Supervised learning (II) -- Multi-Layer Perceptron (MLP) (non-linear regression/function approximation problems) 6 Programming assignment #1 (Due / Oral presentation) 7 Supervised learning (III) -- MLP (scikit-learn) 8 Supervised learning (IV) -- MLP (scikit-learn) (Programming assignment #2) **9 Mid-term paper reading report (Due / Oral presentation--Part I) (預計兩周) 10 Mid-term paper reading report (Due / Oral presentation--Part II) 11 Supervised learning (V) -- MLP (scikit-learn) (time-series data modeling and forecasting) 12 Unsupervised learning -- k-Means clustering (scikit-learn) (PART III) 13 Programming assignment #2 (Due / Oral presentation) 14 Supervised learning (I) -- MLP as a Deep Neural Network (TensorFlow/Keras (GPU accelerated)) 15 Supervised learning (II) -- Convolutional Neural Networks (CNNs) (TensorFlow/Keras (GPU accelerated)) 16 Supervised learning (III) -- Long Short-Term Memory (LSTM) / Supervised/Structured learning (V) -- Generative Adversarial Networks (Generative AI) 17 Final paper reading report (Due / Oral presentation--Part I) (預計兩周) **18 Final paper reading report (Due / Oral presentation--Part II)
評量方式與標準	- Class participation: 20% - Assignments (PPT only): 30% (Two or Three programming assignments: No assignment turned in late will be accepted) - Mid-term paper reading report (PPT only): 25% - Final paper reading report (PPT only): 25%
使用教材、參考書目或其他	【遵守智慧財產權觀念，請使用正版教科書，不得使用非法影印教科書】 使用外文原文書：是 Course lectures will be distributed through our University's iStudy+ (e-learning) system.
課程諮詢管道	e-mail: yhlin@ntut.edu.tw LINE : this course's LINE group (will be announced)
課程對應SDGs指標	SDG9：產業創新與基礎設施（Industry, Innovation and Infrastructure） SDG11：永續城市與社區（Sustainable Cities and Communities）
課程是否導入AI
備註	Remote Learning will be performed via Google Meet/Microsoft Teams, if required (see https://news.ntut.edu.tw/p/422-1000-1085.php?Lang=zh-tw).