Hsuan-Tien Lin > Courses > Machine Learning, Fall 2026

Hsuan-Tien Lin

Home | MOOCs | AIsk | Courses | Research Group | Awards | Publications | Presentations | Programs/Data

Machine Learning, Fall 2026

Course Description

Machine learning allows computational systems to adaptively improve their performance with experience accumulated from the data observed. This course introduces the basics of learning theories, the design and analysis of learning algorithms, and some applications of machine learning.

People

instructor: Hsuan-Tien LIN (htlin AT csie . ntu . edu . tw) [office hour: after classes, or by appointment]
TAs and TA hour: html_ta AT csie . ntu . edu . tw
- Cheng-Han Hsieh (Head TA), CSIE
- Guan-Chen Chen, CSIE
- Chun-Yang Cheng, CSIE
- Sheng-Yuan Chang, CSIE
- Tsung-Lun Chiang, EE
- Ting-Chieh Chiu, CSIE
- Ting-Wei Chuang, CSIE
- Kang-Yu Fan, CSIE
- Yen-Hung Hsueh, EE
- Li-Cheng Hsu, EE
- Jung-Sheng Hung, GINM
- Yun-Shi Lee, IM
- Yi-Ting Lo, CSIE
- Han-Hwa Shih, GICE
- Yen-Chu Tsai, CSIE
- Yi-Ming Yao, Math

Course Information

Time: Wednesdays 9:10 to 12:10
Room: CSIE R103 (with broadcast to R102, R104 and screencast link to TAICA below)
NTU COOL: https://cool.ntu.edu.tw/courses/63348
Synchronous Screencast (TAICA and public): here, containing slido interaction at #HTML2026FALL
Textbook: Learning from Data, by Yaser Abu-Mostafa, Malik Magdon-Ismail and Hsuan-Tien Lin
Language: Mandarin teaching
Grading: 30% homework, 30% exam, 40% project (tentative)
NTU Signup Link: here
Auditing Request Link: here

Announcements

2026/09/10: homework 0 announced
2026/09/09: Auditing request link announced here
2026/08/30: NTU signup link announced here
2026/08/29: course policy announced here

Class Policy

policy

Course Plan (tentative)

date	syllabus	todo/done	materials
09/09 (W1)	course introduction; topic 1: when can machines learn?	homework 0 announced	slides for course introduction
09/16 (W2)	the learning problem learning to answer yes/no; types of learning (first half);	homework 1 announced	slides for the learning problem; LFD 1.0, 1.1.1, 1.2.4 slides for learning to answer yes/no; LFD 1.1.2, 3.1 slides for types of learning; LFD 1.2; LFD 1.3
09/23 (W3)	feasibility of learning; topic 2: why can machines learn? training versus testing		slides for feasibility of learning; LFD 1.3 suggested extended reading: The Lack of A Priori Distinctions Between Learning Algorithms (Wolpert) slides for training versus testing; LFD 2.0, 2.1.1
09/30 (W4)	(optional)theory of generalization; the VC dimension; noise and error	homework 2 announced	slides for the VC dimension; LFD 2.2 slides for noise and error; LFD 1.4 optional: slides for theory of generalization ; LFD 2.1.2
10/07 (W5)	topic 3: how can machines learn? linear regression; logistic regression		slides for linear regression; LFD 3.2 slides for logistic regression; LFD 3.3
10/14 (W6)	linear models for classification; nonlinear transformation	homework 0 due; homework 1 due; homework 2 due; homework 3 announced; final project announced (tentative)	slides for linear models for classification; LFD 3.3 slides for nonlinear transformation; LFD 3.4
10/21 (W7)	topic 4: how can machines learn better? hazard of overfitting; regularization		slides for hazard of overfitting; LFD 4.0, 4.1 slides for regularization; LFD 4.2
10/28 (W8)	validation; three learning principles	homework 3 due; homework 4 announced	slides for validation; LFD 4.3 slides for three learning principles; LFD 5
11/04 (W9)	topic 5: how can machines learn by embedding numerous features? linear support vector machine; dual support vector machine		slides for linear support vector machine; LFD e-8.1 slides for dual support vector machine; LFD e-8.2
11/11 (W10)	kernel support vector machine; soft-margin support vector machine	homework 4 due; homework 5 announced	slides for kernel support vector machine; LFD e-8.3 slides for soft-margin support vector machine; LFD e-8.4
11/18 (W11)	topic 6: how can machines learn by combining predictive features? blending and bagging; adaptive boosting		slides for blending and bagging suggested extended reading: A linear ensemble of individual and blended models for music rating prediction (Chen et al.) suggested extended reading: Bagging predictors (Breiman) slides for adaptive boosting suggested extended reading: A short introduction to boosting (Freund and Schapire)
11/25 (W12)	decision tree; random forest; gradient boosted decision tree	homework 5 due; homework 6 announced	slides for decision tree suggested extended reading: Classification and regression trees (overview of decision tree by Loh) suggested extended reading: Classification and regression trees (book of CART by Breiman et al.) slides for random forest suggested extended reading: Random forest (Breiman) slides for gradient boosted decision tree suggested extended reading: Greedy Function Approximation: A Gradient Boosting Machine (Friedman)
12/02 (W13)	no class as instructor needs to attend ACML 2025 and NeurIPS 2026;		slides for machine learning for modern artificial intelligence
12/09 (W14)	FINAL EXAM!!!;
12/16 (W15)	topic 7: how can machines learn by distilling hidden features? neural network; deep learning	homework 6 due	slides for neural network; LFD e-7.1, e-7.2, e-7.3, e-7.4 (selected parts) slides for deep learning; LFD e-7.6
12/23 (W16)	modern deep learning; finale		slides for deep learning activation ReLU: Deep sparse rectifier neural networks (Glorot, Bordes and Bengio) leaky ReLU: Rectifier Nonlinearities Improve Neural Network Acoustic Models (Maas, Hannun and Ng) parametric ReLU: Delving Deep into Rectifiers: Surpassing Human-Level Performance on Image Net Classification (He, Zhang, Ren and Sun) slides for deep learning optimization backprop and momentum: Learning representations by back-propagating errors (Rumelhart, Hinton, and Williams); On the Momentum Term in Gradient Descent Learning Algorithms (Qian) adam: Adam: A Method for Stochastic Optimization (Kingma and Ba) slides for super short summary
12/30 (W17)	no class and winter vacation started (really?)	final project due (tentative)

Last updated at CST 06:19, September 11, 2026
Please feel free to contact me: htlin.email.png