Hsuan-Tien Lin

Home | MOOCs | Courses | Research Group | Awards | Publications | Presentations | Programs


Machine Learning, Fall 2010

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.

Course Information

Announcements

Class Policy

Course Plan (tentative)

datesyllabustodo/donesuggested reading
2010.09.06(last week of summer!) homework 0 released
2010.09.13introduction/perceptrons homework 1 released; policy explained here taught in class:
  • beginning of Ch 1
  • some of Sec 1.1
  • beginning of Sec 1.2
  • Subsec 1.2.1
reading assignment:
  • Sec 1.1
  • Subsec 1.2.2
2010.09.20perceptrons/setup/learnability advantages announced taught in class:
  • Sec 1.3
  • beginning of Sec 1.4
  • some of Subsec 1.4.1
reading assignment:
  • Subsec 1.4.1
2010.09.27learnability/generalization homework 1 due; homework 2 released; sidework 1 on the proof of Hoeffding's inequality released taught in class:
  • Subsec 1.4.1, 1.4.2
  • beginning of Ch 2 and Sec 2.1
reading assignment:
  • Subsec 1.4.3
2010.10.04generalization/VC dimension taught in class:
  • Page 2-4, 2-5, 2-6, 2-7
  • Definition 2.4
reading assignment:
  • Page 2-1 to 2-7
2010.10.11VC inequality/PLR revisited homework 2 due; homework 3 released taught in class:
  • selected parts of Section 2.1
  • Page 3-1, 3-2, 3-3
reading assignment:
  • whole Section 2.1
other suggested material:
  • the note on VC proof given in class
2010.10.18practical VC/linear classification taught in class: reading assignment: other suggested material:
  • Section 2.3, which is yet another proof of the VC bound
2010.10.25linear model homework 3 due; homework 4 released taught in class:
  • selected parts of Sections 3.2 and 3.3
reading assignment:
  • Sections 3.2, 3,3 and 3.4
2010.11.01overfitting/regularization/validation taught in class:
  • selected parts of Sections 4.1, 4.2 and 4.4
  • Subsetion 4.3.1
reading assignment:
  • Sections 3.4, 4.1, 4.2
  • Subsection 4.3.1
2010.11.08validation/support vector machines homework 4 due; homework 5 released taught in class:
  • selected parts of Sections 4.4
  • beginning of Chapter 8 (slides released via class email)
reading assignment:
  • Section 4.4
  • Chapter 5
  • pages 8-1 to 8-12 of the Chapter 8 slides
2010.11.15(no class---NTU anniversary!)
2010.11.22support vector machines sidework 2 on KKT conditions released taught in class:
  • continuing on Chapter 8 (slides released via class email)
reading assignment:
  • pages 8-13 to 8-20 of the Chapter 8 slides
2010.11.29support vector machines/practical tools homework 5 due; final project announced; homework 6 released; taught in class:
  • continuing on Chapter 8 (slides released via class email)
  • practical tools (for final project): scaling, missing values, parameter selection, large-scale data
reading assignment:
  • pages 8-21 and beyond of the Chapter 8 slides
2010.12.06multiclass/aggregation/bagging taught in class:
  • multiclass classification via binary classification and regression (check the external slides)
  • beginning of Chapter 10 (slides released via class email)
reading assignment:
  • pages 10-1 to 10-11 of the Chapter 10 slides
2010.12.13boosting/decision trees homework 6 due; homework 7 released taught in class: reading assignment:
  • pages 10-12 to 10-23 of the Chapter 10 slides
2010.12.20decision trees/random forest/feature selection taught in class: reading assignment:
  • pages 10-23 to 10-29 of the Chapter 10 slides
2010.12.27Bayesian approaches homework 7 due; optional homework 8 released taught in class:
2011.01.03Bayesian approaches homework 8 due taught in class:
  • Gaussian Discriminant Analysis and its connection to Linear Classification
  • Naive Bayes Classifier
  • contrasting Bayesian with other approaches
2011.01.10summary/final project discussions final project due taught in class:

Last updated at CST 01:20, February 17, 2015
Please feel free to contact me: htlin.email.png
Valid HTML 4.0!