Computer Engineering, Pattern Recognition and Machine Learning, 7.5 credits
Please note that the literature can be changed/revised until:
• June 1 for a course that starts in the autumn semester
• November 15 for a course that starts in the spring semester
• April 1 for a course that starts in the summer
Print or save the syllabus as a PDF
You can easily print a syllabus from the website. Use the keyboard shortcut ctrl+p (Windows) or command+p (Mac). In the next step, you choose whether you want to print or save the course plan as a PDF.
Syllabus:
Datateknik, Mönsterigenkänning och maskininlärning, 7,5 hp
Computer Engineering, Pattern Recognition and Machine Learning, 7.5 credits
General data
- Code: DTA017F
- Subject/Main field: Computer Engineering
- Cycle: Third cycle
- Credits: 7,5
- Answerable faculty: Faculty of Science, Technology and Media
- Answerable department: Computer and Electrical Engineering
- Approved: 2023-05-26
- Version valid from: 2022-01-01
Aim
The doctoral student should understanding machine learning techniques. The doctoral student should develop skills in finding interesting features, building deep learning models and implement deep learning models. The doctoral student should show an ability to apply the skills in a research project.
Course objectives
Upon completion of the course the student should be able to:
- Describe theoretical underpinnings of pattern recognition and machine learning.
- Explain the fundamental problems of deep learning
- Describe the standard cost functions
- Describe regularization methods
- Describe basic optimization methods for deep learning
- Apply these techniques in a research project,
- Implement several types of machine learning methods and modify them
- Critically evaluate the methods’ applicability in new contexts.
- Analyse deep learning in state-of-the-art research within the student's own research field.
Content
The course includes the following elements:
• Ensemble methods
• Multilayer perceptron
• Convolutional neural network
• Recurrent neural network
• Regularization for deep Learning
• Optimization for training deep models
• Implementing deep Learning models
• Graphic models
• Autoencoders
Entry requirements
A person meets the entry requirements for the course if he or she has been admitted to a third-cycle study programme and will be given credit for the course in that study programme.
(Äldre gymnasiebetyg)
Selection rules and procedures
The selection process is in accordance with the Higher Education Ordinance and the local order of admission.
Teaching form
Teaching is carried out by means of the following elements:
- Lectures,
- Assignments,
- Practical tasks performed individually,
- Oral presentations of completed tasks individually ,
- Written reports of the tasks carried out individually.
Examination form
Examination consists of four parts: participating in seminars, completed exercises, oral presentation of literature study, and a written project report.
Grading system
Fail (U) or Pass (G)
Course reading
Required literature
- Author: Ian H. Witten, Eibe Frank and Mark A. Hall
- Title: Data Mining, Pratical Machine Learning Tools and Techniques
- Edition: 3
- Publisher: MORGAN KAUFMANN
- Comment: ISBN-13: 978-0123748560, ISBN-10: 0123748569
Reference literature
- Author: James, G., Witten, D., Hastie, T., Tibshirani, R.
- Title: An introduction to statistical learning with applications in R
- Publisher: Springer
- Comment: ISBN 978-1-4614-7137-0