Deep Learning

Postgraduate course

Course description

Objectives and Content

Artificial neural networks are flexible and powerful machine learning models. Modern deep learning has had tremendous success in applying complex neural networks to problems from a wide range of disciplines. This course gives and understanding of the theoretical basis underlying neural networks and deep learning. Furthermore, the course includes implementation of neural components and as well as applying deep learning on real-world data sets using modern deep learning packages.

Learning Outcomes

Upon completion of the course the student should have the following learning outcomes defined in terms of knowledge, skills and general competence:

Knowledge
The student should be able to

  • explain the basic principles behind neural networks and deep learning
  • compare modeling aspects of various neural network architectures

Skills
The student should be able to

  • implement simple neural network algorithms
  • apply and evaluate deep learning on real data sets

General competence
The student should be able to

  • provide successful examples how deep learning can be used in different contexts in the society
  • read and critically assess papers on artificial neural networks and their applications

Level of Study

Master

Semester of Instruction

Spring
Required Previous Knowledge
For incoming exchange students: At least 60 ECTS in Computer Science and at least 10 ECTS in mathematics
Recommended Previous Knowledge
Machine learning, INF264 or equivalent.
Programming skills, INF102 or equivalent
Good mathematical background, especially linear algebra, calculus and probability (e.g MAT111, MAT121, STAT110)
Credit Reduction due to Course Overlap
None
Access to the Course
Access to the course requires admission to a programme of study at The Faculty of Mathematics and Natural Sciences.
Teaching and learning methods

Lectures, mac 4 hours per week

Exercises, 2 hours per week

Independent projects

Compulsory Assignments and Attendance
Approved compulsory exercises. Compulsory assignments are valid for two semester; the semester the assignments were conducted and the subsequent one.
Forms of Assessment
Written examination (3 hours). The compulsory exercises can be graded and this grade can count for the final grade. Both the exam and the compulsory exercises must be passed.
Grading Scale
The grading scale used is A to F. Grade A is the highest passing grade in the grading scale, grade F is a fail.
Assessment Semester
Examination both spring semester and autumn semester. In semesters without teaching the examination will be arranged at the beginning of the semester.
Reading List
The reading list will be available within June 1st for the autumn semester and December 1st for the spring semester.
Course Evaluation
The course will be evaluated by the students in accordance with the quality assurance system at UiB and the department.
Examination Support Material
None
Programme Committee
The Programme Committee is responsible for the content, structure and quality of the study programme and courses.
Course Coordinator
Course coordinator and administrative contact person can be found on Mitt UiB, or contact Student adviser
Course Administrator
The Faculty of Mathematics and Natural Sciences represented by the Department of Informatics is the course administrator for the course and study programme.