機器學習概論
Introduction to Machine Learning
| 節 | 週二 |
|---|---|
5 13:20–14:10 | 機器學習概論 EC122（光復） 3 節連堂 |
6 14:20–15:10 | |
7 15:30–16:20 |
* 根據陽明交大上課時間表所列
本課程為大學部人工智慧與機器學習相關進階課程之基礎課程，目標在於建立學生對機器學習核心概念、常用模型與現代 AI 發展脈絡的整體理解。課程主軸為「表徵」：如何將資料表示成適合機器學習模型處理的形式。課程內容兼顧傳統機器學習與深度學習方法，比例約為 1:1。傳統機器學習部分涵蓋 regression、classification、model evaluation、regularization、decision trees、ensemble methods、SVM、kernel methods、dimensionality reduction 與 clustering；深度學習部分則涵蓋 neural networks、CNNs、sequence models、transformers、generative models、diffusion models 與 multimodal applications。 本課程不以深入數學推導或進階模型細節為主要目標，而是強調概念理解、模型使用情境、基本實作能力、模型評估方法，以及不同方法之間的關聯。修課學生將透過 Python 與現代機器學習框架實作、評估並應用常見機器學習模型。較深入的理論、模型架構與研究議題，將由後續人工智慧、深度學習、電腦視覺、自然語言處理、大型語言模型與相關進階課程承接。 This course is designed as a foundational undergraduate course for students who plan to take advanced courses in artificial intelligence, computer vision, natural language processing, deep learning, and related areas. The central theme of the course is representation: how data can be represented in forms that are suitable for machine learning models. The course covers both classical machine learning and modern deep learning, with an approximate balance between the two. Classical topics include supervised learning, model evaluation, regularization, decision trees, ensemble methods, support vector machines, kernel methods, dimensionality reduction, and clustering. Deep learning topics include neural networks, CNNs, sequence models, transformers, generative models, diffusion models, and multimodal applications. The course emphasizes conceptual understanding, model usage, practical implementation, evaluation, and connections among different learning methods. It is not intended to focus primarily on advanced mathematical derivations or specialized model details. More advanced theoretical, architectural, and research-oriented topics will be further developed in subsequent courses such as Artificial Intelligence, Deep Learning, Computer Vision, Natural Language Processing, Large Language Models, and related graduate-level courses.
線性代數、機率與統計、微積分、Python 程式設計能力。具備 PyTorch、TensorFlow 或 Keras 等機器學習／深度學習框架的基本使用經驗，以及資料前處理經驗者尤佳，但非必要。 Students are expected to have prior coursework in linear algebra, probability and statistics, and calculus, as well as basic programming skills in Python. Familiarity with machine learning frameworks such as PyTorch, TensorFlow, or Keras, and basic data preprocessing is recommended but not strictly required.
無備註
To be announced.
Homework: 25% Zuvio / quizzes / attendance: 15% Midterm Exam 1 (Oct 13): 20% Midterm Exam 2 (Nov 24): 20% Final Project (Dec 15 and 22): 20%
教師未提供此項資料
| 週次 | 主題 |
|---|---|
| 第 1 週 | Introduction, Representation, ML Workflow 2026-09-08(二) 時數：[2026-09-08]陳昱芝(3.00) |
| 第 2 週 | Linear Regression and Gradient Descent 2026-09-15(二) 時數：[2026-09-15]陳昱芝(3.00) |
| 第 3 週 | Logistic Regression and Classification 2026-09-22(二) 時數：[2026-09-22]陳昱芝(3.00) |
| 第 4 週 | Bias–Variance, Regularization and Model Evaluation 2026-09-29(二) 時數：[2026-09-29]陳昱芝(3.00) |
| 第 5 週 | Decision Trees and Ensemble Learning 2026-10-06(二) 時數：[2026-10-06]陳昱芝(3.00) |
| 第 6 週 | Midterm 1 + Review 2026-10-13(二) 時數：[2026-10-13]陳昱芝(3.00) |
| 第 7 週 | Kernel Methods and SVM 2026-10-20(二) 時數：[2026-10-20]陳昱芝(3.00) |
| 第 8 週 | Dimensionality Reduction and Clustering 2026-10-27(二) 時數：[2026-10-27]陳昱芝(3.00) |
| 第 9 週 | Neural Networks and Representation Learning 2026-11-03(二) 時數：[2026-11-03]陳昱芝(3.00) |
| 第 10 週 | CNN and Transfer Learning 2026-11-10(二) 時數：[2026-11-10]陳昱芝(3.00) |
| 第 11 週 | Sequence Models: RNN/LSTM → Attention 2026-11-17(二) 時數：[2026-11-17]陳昱芝(3.00) |
| 第 12 週 | Midterm 2 + Transformer Overview 2026-11-24(二) 時數：[2026-11-24]陳昱芝(3.00) |
| 第 13 週 | Generative Models: Autoencoder, GAN, Diffusion 2026-12-01(二) 時數：[2026-12-01]陳昱芝(3.00) |
| 第 14 週 | Multimodal AI: CLIP, VLM, LLM Applications 2026-12-08(二) 時數：[2026-12-08]陳昱芝(3.00) |
| 第 15 週 | Final Project Presentation I 2026-12-15(二) 時數：[2026-12-15]陳昱芝(3.00) |
| 第 16 週 | Final Project Presentation II + ML Roadmap 2026-12-22(二) 時數：[2026-12-22]陳昱芝(3.00) |
1. C. Bishop, Pattern Recognition and Machine Learning, Springer 2006 https://www.springer.com/gp/book/9780387310732 Free pdf download: https://www.microsoft.com/en-us/research/uploads/prod/2006/01/Bishop-Pattern-Recognition-and-Machine-Learning-2006.pdf 2. Ian Goodfellow, Yoshua Bengio, and Aaron Courville, Deep Learning, MIT Press, 2016 Free pdf download: https://www.deeplearningbook.org/ 3. 李宏毅深度學習教學LeeDL-Tutorial（蘋果書） Free pdf download: https://github.com/datawhalechina/leedl-tutorial
- 地點
- EC241B (my office)
- 時間
- 4:20~5:20 pm on Tuesdays at EC241B (my office); other time slots: email me first
- 聯絡方式
- Email: berriechen@nycu.edu.tw