Home > Programme > Courses Offered
Course Information
COMP3354 Statistical Learning

### COMP3354 Statistical Learning

2019-20
Instructor(s):Luo Hao
(Class A) No. of credit(s):6
Recommended Learning Hours:
 Lecture: 26 Tutorial: 13
Pre-requisite(s):MATH1853 or MATH2101 or STAT1602 or STAT1603
Co-requisite(s):
Mutually exclusive with:
Remarks:

Course Learning Outcomes

 1 [1] demonstrate an understanding of the conceptual underpinnings of various statistical learning techniques in terms of how, why and when each method works in different real-life scenarios 2 [2] critically evaluate the analytical strategies adopted in applying statistical learning techniques to different areas 3 [3] apply basic statistical learning methods to perform exploratory data analysis and build predictive models using the statistical programming environment R, with real datasets 4 [4] properly tune, select, and validate statistical learning models 5 [5] interpret the results and discuss their implication
Mapping from Course Learning Outcomes to Programme Learning Outcomes
PLO aPLO bPLO cPLO dPLO ePLO fPLO gPLO hPLO iPLO j
CLO 1TT
CLO 2TTT
CLO 3TT
CLO 4TT
CLO 5TT

T - Teach, P - Practice
For BEng(CompSc) Programme Learning Outcomes, please refer to here.

Syllabus

Calendar Entry:
The challenges in learning from big and complicated data have led to significant advancements in the statistical sciences. This course introduces supervised and unsupervised learning, with emphases on the theoretical underpinnings and on applications in the statistical programming environment R. Topics include linear methods for regression and classification, model selection, model averaging, basic expansions and regularization, kernel smoothing methods, additive models and tree-based methods. We will also provide an overview of neural networks and random forests.

Detailed Description:

Introduction Mapped to CLOs
Overview of supervised learning1, 2
Supervised learning Mapped to CLOs
Linear methods for regression and classification1, 2, 3, 5
Basic expansions and regularization1, 2, 3, 5
Kernel smoothing methods1, 2, 3, 5
Model assessment and selection1, 2, 3, 4, 5
Model inference and averaging1, 2, 3, 4, 5
Additive models and tree-based methods1, 2, 3, 5
Unsupervised learning Mapped to CLOs
Cluster analysis and principal component analysis1, 2, 3, 5
Advanced topics and applications Mapped to CLOs
Neural networks1, 2, 3, 5
Random forests1, 2, 3, 5
Applications1, 2, 3, 4, 5

Assessment:
Continuous Assessment: 50%
Written Examination: 50%

Teaching Plan

Please refer to the corresponding Moodle course.

Moodle Course(s)