Machine Learning 1

Course manual 2019/2020

Course content

This course is lecture based, with homework exercises and programming assignments.

The curriculum is based on the chapters 1, 2, 3, 4, 5, 6, 7, 9, 14 of the book "Pattern Recognition and Machine Learning" by C. Bishop:

• Statistical learning principles
• Linear regression
• Linear classification
• Neural networks
• Kernel methods
• Dimensionality reduction
• Clustering methods
• Ensemble methods

Study materials

Literature

• C.M. Bishop, 'Pattern Recognition and Machine Learning', 2006, Springer, ISBN 0-38-731073-8

Software

• Python

Objectives

• The student can explain, motivate and distinguish the main areas of machine learning in general and on examples.
• The student can explain the major statistical learning frameworks/principles together with their advantages and shortcomings.
• The student knows the major linear and non-linear statistical models together with their advantages and disadvantages, can explain them and the made model assumptions, can reason about and inside them, and can manipulate them.
• The student can set up learning objectives with the taught models, can train and evaluate them and can assess the quality of fit.
• The student can implement all the above in Python and apply the learned principles and models to real world problems and data sets.

Teaching methods

• Lecture
• Computer lab session/practical training
• Self-study
• Discussion forum
• Exercise group
• Homework exercise
• Programming assignment
• Practice exercise

In the lectures the new concepts are introduced. The direct interaction with the teacher allows for questions, clarification and immediate feedback.

The ungraded practice exercises with solutions allow the students to test, train and deepen their understanding of the new learned concepts with the possibility for direct feedback via self-checks.

The graded homework exercises closely follow the practice exercises but also aim at assessing the students' progress.

The programming assignments trains the students ability to implement the learned concepts and apply them to real world problems. The direct feedback loops via the computer helps the students to self correct quickly. The computer assignments are also graded and assess the students' skills.

The exercise groups, computer labs and discussion forum are places where the students can voice questions, get answers and feedback from peers and TAs.

Learning activities

Attendance

The programme does not have requirements concerning attendance (OER-B).

Assessment

To pass the course the student needs to have at least 50% of the points in the final exam (Tentamen).

Lab assignments and homework exercises will not be graded if they are handed in after their deadlines.

The grade of the resit exam will replace 100% of all the grades in the course (final exam, lab assignments, homework exercises).

Assignments

Practice exercises

• ungraded

Homework exercises

• graded; 5-6 exercise sheets in total; to be made individually

Lab assignments

• graded; 3 assignments in total; to be handed in in pairs

Lab assignments and homework exercises will not be graded if they are handed in after their deadlines.

Feedback is given in the exercise groups, computer lab sessions and via the discussion forum.

Fraud and plagiarism

The 'Regulations governing fraud and plagiarism for UvA students' applies to this course. This will be monitored carefully. Upon suspicion of fraud or plagiarism the Examinations Board of the programme will be informed. For the 'Regulations governing fraud and plagiarism for UvA students' see: www.student.uva.nl

Course structure

Timetable

The schedule for this course is published on DataNose.

Additional information

Recommended prior knowledge: probability theory, statistics, linear algebra, (vector) calculus, programming.

Contact information

Coordinator

• dr. P.D. Forré

For all information regarding the assignments please contact your TA or find help in the discussion forum.