COMP SCI 7401 - Introduction to Statistical Machine Learning

North Terrace Campus - Semester 2 - 2014

Statistical Machine Learning is concerned with algorithms that automatically improve their performance through "learning". For example, computer programs that learn to detect humans in images/video; predict stock markets, and rank web pages. Statistical machine learning has emerged mainly from computer science and artificial intelligence, and has connections to a variety of related subjects including statistics, applied mathematics and pattern analysis. Applications include image and audio signal analysis, data mining, bioinformatics and exploratory data analysis in natural science and engineering. This is an introductory course on statistical machine learning which presents an overview of many fundamental concepts, popular techniques, and algorithms in statistical machine learning. It covers basic topics such as dimensionality reduction, linear classification and regression as well as more recent topics such as ensemble learning/boosting, support vector machines, kernel methods and manifold learning. This course will provide the students the basic ideas and intuition behind modern statistical machine learning methods. After studying this course, students will understand how, why, and when machine learning works on practical problems.

  • General Course Information
    Course Details
    Course Code COMP SCI 7401
    Course Introduction to Statistical Machine Learning
    Coordinating Unit Computer Science
    Term Semester 2
    Level Postgraduate Coursework
    Location/s North Terrace Campus
    Units 3
    Contact Up to 2 hours per week
    Prerequisites COMP SCI 7201
    Assumed Knowledge Basic probability theory and linear algebra
    Course Description Statistical Machine Learning is concerned with algorithms that automatically improve their performance through "learning". For example, computer programs that learn to detect humans in images/video; predict stock markets, and rank web pages. Statistical machine learning has emerged mainly from computer science and artificial intelligence, and has connections to a variety of related subjects including statistics, applied mathematics and pattern analysis. Applications include image and audio signal analysis, data mining, bioinformatics and exploratory data analysis in natural science and engineering. This is an introductory course on statistical machine learning which presents an overview of many fundamental concepts, popular techniques, and algorithms in statistical machine learning. It covers basic topics such as dimensionality reduction, linear classification and regression as well as more recent topics such as ensemble learning/boosting, support vector machines, kernel methods and manifold learning. This course will provide the students the basic ideas and intuition behind modern statistical machine learning methods. After studying this course, students will understand how, why, and when machine learning works on practical problems.
    Course Staff

    Course Coordinator: Professor Javen Qinfeng Shi

    Machine Learning is the process of using data to uncover an underlying process. It is used in a wide range of applications including stock market prediction, fraud detection, recommendation system, social networks, medical diagnosis, security and so on. Big IT companies like Google, Amazon, Facebook, Microsoft which have large demand for hiring graduates with machine learning background and experience.

    In this course, we will cover the basic concepts of machine learning, and classic algorithms such as Support Vector Machines, as well as more up-to-date tools such as Probabilistic Graphical Models (PGMs). Apart from the algorithms, we will also cover the principles and theory of machine learning, which could guide the students to analyse existing algorithms, and to design or invent their own algorithms in future.

    We will have a mixture of hands-on programming practice and mathematical derivation in terms of assignments. We hope the experience and knowledge from this course can broaden students’ career options such as doing a PhD in relevant areas, or getting a better chance into relevant industry in the Big Data era. 


    The course will be taught by 
     1. Dr. Qinfeng (Javen) Shi (Course Coordinator)
             2. Prof. Chunhua Shen

    Dr Qinfeng (Javen) Shi is an ARC DECRA fellow with expertise in Machine Learning. More info can be found in http://cs.adelaide.edu.au/~javen/.

    Prof. Chunhua Shen is an ARC future fellow with expertise in Machine Learning and Computer Vision. More info can be found in http://cs.adelaide.edu.au/~chhshen/.
    Course Timetable

    The full timetable of all activities for this course can be accessed from Course Planner.

    The full timetable of all activities for this course can be accessed from Course Planner.Teachers:

    Dr. Qinfeng (Javen) Shi (Course Coordinator)

    Prof. Chunhua Shen


    Lecture Time:
    28 Jul. - 15 Sep.                     Monday          2pm - 4pm     
    6 Oct. - 27 Oct.                       Monday          2pm - 4pm     

     
    Location:                   Ingkarni Wardli, B20, Teaching Suite


    Schedule:
    28 July  [Javen] Machine Learning Problem
    4 Aug.   [Javen] Supervised Learning --- Classification --- 
                            Algorithms including: KNN, Perceptron, SVM etc.
    11 Aug. [Chunhua] Supervised Learning --- Classification ---
                                  Boosting Algorithms
    18 Aug. [Chunhua] Supervised Learning --- Regression
    25 Aug. [Chunhua] Unsupervised Learning and Semi-Supervised Learning
    1 Sep.   [Chunhua] Dimension reduction
    8 Sep.   [Javen] Probabilistic Graphical Models (PGMs) --- Representation
    15 Sep. [Javen] PGMs --- Inference
     
    Mid-break
     
    6 Oct.    [Javen] PGMs --- Learning 
    13 Oct.  [Javen] PGMs --- Structure Estimation
    20 Oct.  [Javen] Kernels
    27 Oct.  [Javen] Learning Theory
  • Learning Outcomes
    Course Learning Outcomes
    (1) Understanding of basic concepts of machine learning, and classic algorithms such as Support Vector Machines, as well as more up-to-date tools such as Probabilistic Graphical Models (PGMs).
    (2) Understanding of basic principles and theory of machine learning, which may guide students to invent their own algorithms in future.
    (3) Ability to program the algorithms in the course.
    (4) Ability to do mathematical derivation of the algorithms in the course.
    University Graduate Attributes

    This course will provide students with an opportunity to develop the Graduate Attribute(s) specified below:

    University Graduate Attribute Course Learning Outcome(s)
    Knowledge and understanding of the content and techniques of a chosen discipline at advanced levels that are internationally recognised. 1
    The ability to locate, analyse, evaluate and synthesise information from a wide variety of sources in a planned and timely manner. 3,4
    A proficiency in the appropriate use of contemporary technologies. 1
    A commitment to continuous learning and the capacity to maintain intellectual curiosity throughout life. 2
  • Learning Resources
    Required Resources
    • No textbook required.
    • Knowing some basic statistics, probability, linear algebra and optimisation would be helpful, but not essential.
    • They will be covered when needed. Ability to program in Matlab, C/C++ is required. 
    Recommended Resources
    Recommended books:
    1. Pattern Recognition and Machine Learning by Bishop, Christopher M.
    2. Kernel Methods for Pattern Analysis by John Shawe-Taylor, Nello Cristianini.
    3. Convex Optimization by Stephen Boyd and Lieven Vandenberghe.
    Book 1 is for machine learning in general. Book 2 focuses on kernel methods with pseudo code and some theoritical analysis. Book 3 gives introduction to (Convex) Optimization.
    Online Learning
    Our course forum is accessible via: http://forums.cs.adelaide.edu.au/


    Excellent external courses available online:
    1. Learning from the data by Yaser Abu-Mostafa in Caltech.
    2. Machine Learning by Andrew Ng in Stanford.
    3. Machine Learning (or related courses) by Nando de Freitas in UBC. 
  • Learning & Teaching Activities
    Learning & Teaching Modes
    The course will be primarily delivered through two activities:
    • Lectures
    • Assignments
    Lectures will introduce and motivate the basic concepts of each topic. Significant discussions and two-way communication are also expected during lectures to enrich the learning experience. The assignments will reinforce concepts by their application to problem solving. This will be done via programming work and mathmatical derivation. All material covered in the lectures and assignments are assessable.
    Workload

    The information below is provided as a guide to assist students in engaging appropriately with the course requirements.

    This is a 3-unit course. Students are expected to spend about 8 hours per week on the course. This includes a 2-hour lecture, 2-hour self study and up to 4 hours per week on completing assignments.

    Assigmment work will be subjected to deadlines. Students are expected to manage their time effectively to allow timely submission, especially with consideration to workload of other courses.
    Learning Activities Summary
    Students are encouraged to attend lectures as material presented in lectures often includes more than is on the slides. Students are also encouraged to ask questions during the lectures. Slides will be available via the subject web page.
    Specific Course Requirements
    • Knowing some basic statistics, probability, linear algebra and optimisation would be helpful, but not essential. They will be covered when needed.
    • Ability to program in Matlab, C/C++ is required. 
  • Assessment

    The University's policy on Assessment for Coursework Programs is based on the following four principles:

    1. Assessment must encourage and reinforce learning.
    2. Assessment must enable robust and fair judgements about student performance.
    3. Assessment practices must be fair and equitable to students and give them the opportunity to demonstrate what they have learned.
    4. Assessment must maintain academic standards.

    Assessment Summary
    The course includes the following assessment components:
    • Final written exam at 55%.
    • Three assignments at 15% each.
    Assessment Related Requirements
    Students must obtain at least 50% of the overall marks to pass the course. Passing is also subjected to meeting the minimum performance hurdle, which is 40% of each course component (final exam and combined assignments).
    Assessment Detail
    Final written exam 
    This will be a 2-hour exam at the end of the course/semester. The exam will assess your knowledge and understanding of the course topics, as well as the abiliity to use the knowledge for problem solving. The exam is open-book. Books, lecture notes, slides print-out, calculators and paper dictionaries (English to foreign language) are permitted. The use of internet is not permitted. 

    Assignments
    Each student is expected to complete assignments in the form of report and programming work. The assignments must be completed individually and all submissions are to be made under the declaration of adherring to the academic honesty principles. Submissions will be subjected to plagiarism checks. This course has a zero-tolerance policy towards academic honesty violations. Offenders will be duly subjected to university procedures for dealing with academic honesty cases.
    Submission
    Assignment
    Assignment solutions are to be submitted through the School of Computer Science's Moodle forum: http://forums.cs.adelaide.edu.au/course/ 
    No physical submissions of work will be accepted unless specifically requested by the lecturer.

    Marks will be capped for late submissions, based on the following schedule:
    1 day late – mark capped at 75%
    2 days late – mark capped at 50%
    3 days late – mark capped at 25%
    more than 3 days late – no marks available.

    Extensions to due dates will only be considered under exceptional medical or personal conditions and will not be granted on the last day due, or retrospectively. Applications for extensions must be made to the course coordinator by e-mail or hard copy and must include supporting documentation – medical certificate or letter from the student counselling service.

    Examination
    The examinations office will schedule the final exam. Students are expected to be available until after the supplementary examination period (precise dates are available from university calendar or exams office). No additional arrangments will be given if students are offered supplementary exams but are unable to attend.
    Course Grading

    Grades for your performance in this course will be awarded in accordance with the following scheme:

    M10 (Coursework Mark Scheme)
    Grade Mark Description
    FNS   Fail No Submission
    F 1-49 Fail
    P 50-64 Pass
    C 65-74 Credit
    D 75-84 Distinction
    HD 85-100 High Distinction
    CN   Continuing
    NFE   No Formal Examination
    RP   Result Pending

    Further details of the grades/results can be obtained from Examinations.

    Grade Descriptors are available which provide a general guide to the standard of work that is expected at each grade level. More information at Assessment for Coursework Programs.

    Final results for this course will be made available through Access Adelaide.

  • Student Feedback

    The University places a high priority on approaches to learning and teaching that enhance the student experience. Feedback is sought from students in a variety of ways including on-going engagement with staff, the use of online discussion boards and the use of Student Experience of Learning and Teaching (SELT) surveys as well as GOS surveys and Program reviews.

    SELTs are an important source of information to inform individual teaching practice, decisions about teaching duties, and course and program curriculum design. They enable the University to assess how effectively its learning environments and teaching practices facilitate student engagement and learning outcomes. Under the current SELT Policy (http://www.adelaide.edu.au/policies/101/) course SELTs are mandated and must be conducted at the conclusion of each term/semester/trimester for every course offering. Feedback on issues raised through course SELT surveys is made available to enrolled students through various resources (e.g. MyUni). In addition aggregated course SELT data is available.

  • Student Support
  • Policies & Guidelines
  • Fraud Awareness

    Students are reminded that in order to maintain the academic integrity of all programs and courses, the university has a zero-tolerance approach to students offering money or significant value goods or services to any staff member who is involved in their teaching or assessment. Students offering lecturers or tutors or professional staff anything more than a small token of appreciation is totally unacceptable, in any circumstances. Staff members are obliged to report all such incidents to their supervisor/manager, who will refer them for action under the university's student’s disciplinary procedures.

The University of Adelaide is committed to regular reviews of the courses and programs it offers to students. The University of Adelaide therefore reserves the right to discontinue or vary programs and courses without notice. Please read the important information contained in the disclaimer.