Hsuan-Tien Lin > Courses > Machine Learning Techniques, Spring 2020

Hsuan-Tien Lin

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

Machine Learning Techniques, Spring 2020

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 (starting on 04/09/2020 Thursday): mltech_ta AT csie . ntu . edu . tw
- Sheng-Feng Wu: Tuesdays 10:00--11:00 Online
- Ching-Yuan Pai: Wednesdays 9:00--10:00, Online
- Si-An Chen: Wednesdays 14:00--15:00, CSIE R536
- Chien-Ming Yu: Thursdays 9:00--10:00, Online
The meeting address of online TA hour is posted in NTU COOL Course Forum.

Course Information

Time: Fridays 10:20 to 12:10
Room: CSIE Building, R103
Textbook: Learning from Data, by Yaser Abu-Mostafa, Malik Magdon-Ismail and Hsuan-Tien Lin
Language: Mandarin teaching with English slides
Grading: 70% homework, 30% project (tentative)
CEIBA Course Website: https://ceiba.ntu.edu.tw/1082mltech
NTU COOL Discussion Forum: https://cool.ntu.edu.tw/courses/1893
screencast link: https://www.csie.ntu.edu.tw/~htlin/mltech20spring/screencast.php

Announcements

homework 3 announced on 05/29/2020
homework 2 announced on 04/24/2020
final project announced on 03/29/2020
homework 1 announced on 03/27/2020
screencast link is here
sign up here if you want to enroll
policy announced on 01/27/2020

Class Policy

main policy

Course Plan (tentative)

date	syllabus	todo/done	suggested reading
3/6	topic 1: how can machines learn by embedding numerous features? linear support vector machine		youtube screencast; course slides; LFD e-8.1
3/13	dual support vector machine		youtube screencast; course slides; LFD e-8.2
3/20	kernel support vector machine		youtube screencast 1; youtube screencast 2; course slides; LFD e-8.3
3/27	soft-margin support vector machine / kernel logistic regression	homework 1 announced; final project announced	youtube screencast; course slides; LFD e-8.4
4/3	no class because of Spring Break
4/10	kernel logistic regression topic 3: how can machines learn by distilling hidden features? neural network		youtube screencast; course slides; extended reading: Kernel Logistic Regression and the Import Vector Machine (Zhu and Hastie) A Note on Platt's Probabilistic Outputs for Support Vector Machines (Lin, Weng and Lin)
4/17	neural network		youtube screencast; course slides; LFD e-7.1, e-7.2, e-7.3, e-7.4 (selected parts)
4/24	(shallow explanation of) deep learning	homework 1 due; homework 2 announced	youtube screencast; course slides; LFD e-7.6
5/1	convolutional neural networks		youtube screencast; skeleton slides; in-class doodle; extended reading: CNN: Backpropagation Applied to Handwritten Zip Code Recognition (LeCun et al.) Dive into Deep Learning Chapter 6
5/8	convolutional neural networks		youtube screencast; notes; in-class doodle; extended reading: CNN: Backpropagation Applied to Handwritten Zip Code Recognition (LeCun et al.) Dive into Deep Learning Chapter 6
5/15	activation in deep learning		youtube screencast; notes; in-class doodle; extended reading: ReLU: Deep sparse rectifier neural networks (Glorot, Bordes and Bengio) leaky ReLU: Rectifier Nonlinearities Improve Neural Network Acoustic Models (Maas, Hannun and Ng) Dive into Deep Learning Section 4.1.2
5/22	initialization / optimization in deep learning	homework 2 due	youtube screencast; notes; in-class doodle; extended reading: Gloret initialization: Understanding the difficulty of training deep feedforward neural networks (Gloret and Bengio) He initialization: Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classification (He et al.) 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) Dive into Deep Learning Sections 4.8, 11.6, 11.8, 11.10
5/29	regularization in deep learning / aggregation	homework 3 announced	youtube screencast; notes on deep learning (in the ones last week); in-class doodle; course slides on aggregation;
6/5	blending / bagging / adaptive boosting		youtube screencast; course slides; extended reading: A linear ensemble of individual and blended models for music rating prediction (Chen et al.) Bagging predictors (Breiman) course slides; extended reading: A short introduction to boosting (Freund and Schapire)
6/12	gradient boosted decision tree / finale		youtube screencast; course slides; extended reading: Greedy Function Approximation: A Gradient Boosting Machine (Friedman) course slides
6/19	no class and good luck with your other finals
6/26	summer vacation started (really?)	final project due

Last updated at CST 13:08, October 04, 2023
Please feel free to contact me: htlin.email.png