The objective of this seminar is to provide students with a solid understanding of machine learning from the Bayesian and optimization perspectives used in data mining problems. By revealing the underlying structure and regularities of the data at hand, which is viewed as learning from data, we developed a model that could be used to provide predictions. Bayesian estimation and optimization theories are very broad concepts and we will focus on issues related to machine learning algorithms. We will examine different learning algorithms using Bayesian estimation and optimization theories.
Students taking this course MUST have the ability to programming in MATLAB or R or other high-level programming languages.
Text:
Reference List: This will be updated during the academic year.
The text/references is/are available at the CUHK library.
The Chinese University of Hong Kong places very high importance on honesty in academic work submitted by students, and adopts a policy of zero tolerance on cheating and plagiarism. Any related offence will lead to disciplinary action including termination of studies at the University. For information on categories of offenses and types of penalties, students should consult the following link: , or doc.
Your final letter-grade will be determined by the following activities as follows:
The seminar, which lasts for about 13 weeks, will be chopped into two sections. We will have an elementary discussion of machine learning algorithms in Section 1. The core of the seminar will be in Section 2. Using Bayesian estimation and optimization techniques, we shall study different machine learning algorithms.
Each student must to present twice: once in Section 1 and once in Section 2.
1. In Section 1, each presentation will last about 2 hours 30 minutes; each topic will be taken from follow the textbook by Sergios Theodoridis. Throughout the 10 weeks and the question and answer interaction discussion, students should have a big picture view of different learning algorithms.
2. In Section 2, students will have at least 20 minutes to summarize her/his report on November 20th, 22th, 27th and 29th, 2018 and 10 minutes for a Q & A session. Each student must submit a full report, no more than 5 pages, on December 6th, 2018 via email as well as a hard-copy pdf/doc file; this report only summarizes your important findings based on your proofs, computations, programming codes and visualization tools, etc. Please attach all the details to the end of your report as the number of pages for that is unlimited.
Once you have enrolled your course, we will send you a username and password to access your online learning resources.