CS565100 Machine Learning, Fall 2015
This course presents a consistent treatment of selected machine learning problems and solutions. Basically, machine learning is about programming computers to optimize a performance criterion using example data or past experience. Consider the recognition of spoken speech—that is, converting the acoustic speech signal to an ASCII text; humans can do this task seemingly without any difficulty, but we are unable to explain how we do it. In machine learning, the approach is to collect a large collection of sample utterances from different people and learn to map these to words. Another example is that developers of a web site (e.g., YouTube) usually collect user behaviors (e.g., mouse clicks), apply machine learning to analyze the preference of individual users, and recommend items (e.g., clips) that may be interesting to these users.
This year we will focus on geometric learning methods. All models and algorithms are explained in deep to help students move from the equations to runnable computer programs.
This course is intended for senior undergraduate and graduate students who have proper understanding of computer programming, calculus, linear algebra, and probability.
Prof. Shan-Hung Wu
Office: Delta 603
Office hour: 13:30 ~ 15:00 on Thursday at Delta 723
* Main contact.
- 2015/12/14 The last TA course is on December 15.
- 2015/11/09 No class on November 10.
- 2015/10/26 No class on October 27.
- 2015/10/19 No class on October 20.
- 2015/10/12 Appendix D "Probability and Statistics" will be covered by TA on Thursday(10/15).
- 2015/10/12 Appendix A "Calculus" will be covered by TA on Tuesday(10/13).
- 2015/09/30 "Convex Optimization" part2 by TA will continue next Tuesday.
- 2015/09/30 Assignment 1 is uploaded.
- 2015/09/16 No class on September 17.
- 2015/09/16 Change classroom to Delta 109.
- 2015/09/16 Course site is online.
- Ethem Alpaydin, Introduction to Machine Learning, 2ed, MIT Press, 2010, ISBN: 026201243X
- Christopher M. Bishop, Pattern Recognition and Machine Learning, Springer, 2006, ISBN: 0387310738
- Ian H. Witten et al., Data Mining: Practical Machine Learning Tools and Techniques, 3ed, Morgan Kaufmann, 2011, ISBN: 0123748569
- 01 Introduction
- 02 Supervised Learning
- 06 Probabilistic Modeling
- 07 Experiments and Ensembling
- 09 Dimensionality Reduction
- 10 Clustering and Expectation Maximization(updated on 11/19)
- 14 Markov Chains and Link Analysis
- 15 Graphical Models(updated on 1/12)
- 16 Hidden Markov Models
- 17 Reinforcement Learning
- A Calculus
- B Linear Algebra and Geometry
- C Convex Optimization
- D Probability and Statistics
- E Information Theory
- Assignment 1 (updated on 2015/10/01) (Solution)
- Assignment 2 (updated on 2015/10/29) (Solution)
- Assignment 3 (updated on 2015/11/19) (Solution)
- Assignment 4 (updated on 2016/1/4) (Solution)
- Midterm Exam: 30%
- Final Exam: 30%
- Assignments & Presentations: 30%
- High-performance rewards: 10%