Course Syllabus

Course Information：

Year-Semester Course Number Course Name Phase Credits Hours Required/Elective Instructor Class Student NO. Withdraw Note

Spring Semester
Year 2018 241591 Machine Learning 1 3.0 3 ★ SHINGCHERN YOU GCSIE
IMEECS 46 15 資工四與資工所、電資外生班合開，修課上限30人

Year-Semester	Course Number	Course Name	Phase	Credits	Hours	Required/Elective	Instructor	Class	Student NO.	Withdraw	Note
Spring Semester Year 2018	241591	Machine Learning	1	3.0	3	★	SHINGCHERN YOU	GCSIE IMEECS	46	15	資工四與資工所、電資外生班合開，修課上限30人

Course Plan：

Instructor SHINGCHERN YOU

Email scyou@ntut.edu.tw

Last Updated 2018-02-27 10:28:28

Course Objective 本課程以深入淺出之方式，介紹與機器學習之相關技術及演算法，目標是修習本課程之學生具有閱讀機器學習相關文獻之基本知識，並能從事相關之實務或電腦模擬實驗之工作，因此本課程著重在廣泛介紹各類之演算法之應用，而非深究少數演算法之深奧數學原理．本課程涵蓋之演算法，包含：最佳化基本知識，隨機最佳化演算法，隨機搜尋，決策樹，隨機森林，類神經網路及深度學習相關演算法，基因演算法，貝氏定理及應用，高斯混合模型，隱藏式馬可夫鍊及訓練演算法，支援向量機，線性及邏輯回歸，分群方法，降維方法，提升方法，決策融合技巧，及增強式學習等．

Course Schedule Week 1: Class announcement and introduction to machine learning (CH 1) Week 2: Basics of spervised learning (Example: K NN), VC dimension, and regression (CH 2). Baysian decision theory (CH 3) and Naive Bayes classifiers (CH 5) Week 3: ML and MAP estimation; Bias vs variance dilemma (CH 4). Sample mean mean and sample covariance (CH 5). Dimension reduction techniques: PCA, FA (CH 6) Week 4: Dimension reduction techniques: LDA and ICA. Clustering algorithm: k-mean (CH 7) Week 5: Unsupervised neural networks (CH 12): Competitive learning, SOFM, and ART. Decision trees: ID3, C4.5, and random forest (CH 8) Week 6: Holiday Week 7: Decision trees (continued). Mixer model: GMM and EM algorithm (CH 7) Week 8: Genetic algorithms. Basics of optiminization. Week 9: MT Week 10: MT sol. Gradient search and linear discrimination (CH 10) Week 11: Feedforward neural networks with examples: radical basis networks and multi-layer perceptrons. Back propagation and regularization methods (CH 11) Week 12: Deep learning and convolutional neural networks Week 13: More on deep neural networks: Autoencoder, LSTM, and others Week 14: SVM (CH 13) and HMM (CH 15) Week 15: Combining multiple classifiers (CH 17) Week 16: Design and analysis of experiments (CH 19) Week 17: Reinforcement learning (CH 18) Week 18: Final exam

Evaluation and grading policy MT 30 % Final 40 % HW 30 % Project 10 % (optional)

Materials All right reserved. Do not reproduce without permission.

Foreign-language textbooks:Yes
Reference text book: Introduction to machine learning. E. Alpaydin. 2nd ed or 3rd ed. Note: textbook is only used to follow the presentation order. Much of the detailed lecture materials are NOT covered in the textbook.

The access to curricular consultation

Note
　

Instructor	SHINGCHERN YOU
Email	scyou@ntut.edu.tw
Last Updated	2018-02-27 10:28:28
Course Objective	本課程以深入淺出之方式，介紹與機器學習之相關技術及演算法，目標是修習本課程之學生具有閱讀機器學習相關文獻之基本知識，並能從事相關之實務或電腦模擬實驗之工作，因此本課程著重在廣泛介紹各類之演算法之應用，而非深究少數演算法之深奧數學原理．本課程涵蓋之演算法，包含：最佳化基本知識，隨機最佳化演算法，隨機搜尋，決策樹，隨機森林，類神經網路及深度學習相關演算法，基因演算法，貝氏定理及應用，高斯混合模型，隱藏式馬可夫鍊及訓練演算法，支援向量機，線性及邏輯回歸，分群方法，降維方法，提升方法，決策融合技巧，及增強式學習等．
Course Schedule	Week 1: Class announcement and introduction to machine learning (CH 1) Week 2: Basics of spervised learning (Example: K NN), VC dimension, and regression (CH 2). Baysian decision theory (CH 3) and Naive Bayes classifiers (CH 5) Week 3: ML and MAP estimation; Bias vs variance dilemma (CH 4). Sample mean mean and sample covariance (CH 5). Dimension reduction techniques: PCA, FA (CH 6) Week 4: Dimension reduction techniques: LDA and ICA. Clustering algorithm: k-mean (CH 7) Week 5: Unsupervised neural networks (CH 12): Competitive learning, SOFM, and ART. Decision trees: ID3, C4.5, and random forest (CH 8) Week 6: Holiday Week 7: Decision trees (continued). Mixer model: GMM and EM algorithm (CH 7) Week 8: Genetic algorithms. Basics of optiminization. Week 9: MT Week 10: MT sol. Gradient search and linear discrimination (CH 10) Week 11: Feedforward neural networks with examples: radical basis networks and multi-layer perceptrons. Back propagation and regularization methods (CH 11) Week 12: Deep learning and convolutional neural networks Week 13: More on deep neural networks: Autoencoder, LSTM, and others Week 14: SVM (CH 13) and HMM (CH 15) Week 15: Combining multiple classifiers (CH 17) Week 16: Design and analysis of experiments (CH 19) Week 17: Reinforcement learning (CH 18) Week 18: Final exam
Evaluation and grading policy	MT 30 % Final 40 % HW 30 % Project 10 % (optional)
Materials	All right reserved. Do not reproduce without permission. Foreign-language textbooks:Yes Reference text book: Introduction to machine learning. E. Alpaydin. 2nd ed or 3rd ed. Note: textbook is only used to follow the presentation order. Much of the detailed lecture materials are NOT covered in the textbook.
The access to curricular consultation
Note