GRA 4160 Advanced Regression and Classification Analysis, Ensemble Methods and Neural Networks

GRA 4160 Advanced Regression and Classification Analysis, Ensemble Methods and Neural Networks

Course code: 
GRA 4160
Department: 
Data Science and Analytics
Credits: 
6
Course coordinator: 
Vegard Høghaug Larsen
Course name in Norwegian: 
Advanced Regression and Classification Analysis, Ensemble Methods and Neural Networks
Product category: 
Master
Portfolio: 
MSc in Data Science for Business
Semester: 
2023 Spring
Active status: 
Active
Level of study: 
Master
Teaching language: 
English
Course type: 
One semester
Introduction

The vast amount of widely available data allows machines to solve challenging tasks without explicitly being programmed to do so, often outperforming existing methods based on domain knowledge and/or human experts.

In this course we focus on basic machine learning methods, both for supervised and unsupervised learning. We will look at how exactly machines "learn" from the data, and how they use the knowledge learned during training to solve tasks of interest.

The course covers both the theory (looking at the methods more rigorously than introductory machine learning courses) and the practice of machine learning (using Python).

Learning outcomes - Knowledge

By the end of the course, the student should be able to:

  • Explain the covered machine learning algorithms,
  • Describe their applications, advantages, disadvantages, and limitations,
  • Choose the appropriate learning methods to solve a given problem of interest.
Learning outcomes - Skills

By the end of the course, the student should be able to:

  • Implement the basic machine learning methods from scratch,
  • Design and build learning systems using existing state-of-the-art machine learning libraries,
  • Train, evaluate, tune, and test these systems,
  • Present and defend the results of their work in a professional and academic manner.
General Competence

During the course, students will also improve their general programming and presentation skills, and practice working in teams.

Course content
  • Recapitulation (supervised vs. unsupervised learning, linear regression and classification).
  • Advanced regression and classification.
  • Clustering.
  • Neural networks (feedforward neural networks, backpropagation algorithm, stochastic gradient descent).
  • Practical aspects of neural networks (e.g., architectures, weights initialization, normalization, regularization, dropout, optimized training methods).
  • Ensemble methods (boosting and bagging).
  • Model selection and assessment.
Teaching and learning activities
  • Organized classes will be a mixture of lectures and working on in-class assignments.
  • Students are expected to prepare for the organized classes by watching selected videos online and/or reading selected texts.
  • Students are also expected to work on assignments outside of the class, either individually or in small groups.
  • Mini project (in groups of 3-4 students).

Software tools: Python (including its state-of-the-art machine learning packages).

Software tools
Software defined under the section "Teaching and learning activities".
Additional information

Please note that while attendance is not compulsory in all courses, it is the student’s own responsibility to obtain any information provided in class.

This is a course with continuous assessment (several exam components) and one final exam code. Each exam component is graded by using points on a scale from 0-100. The components will be weighted together according to the information in the course description in order to calculate the final letter grade for the examination code (course). Students who fail to participate in one/some/all exam elements will get a lower grade or may fail the course. You will find detailed information about the point system and the cut off points with reference to the letter grades when the course starts.

At re-sit all exam components must, as a main rule, be retaken during next scheduled course.

Qualifications

All courses in the Masters programme will assume that students have fulfilled the admission requirements for the programme. In addition, courses in second, third and/or fourth semester can have specific prerequisites and will assume that students have followed normal study progression. For double degree and exchange students, please note that equivalent courses are accepted.

Disclaimer

Deviations in teaching and exams may occur if external conditions or unforeseen events call for this.

Required prerequisite knowledge

Programming (preferably Python), intermediate math (including linear algebra, basic calculus, and basics of theory of probability), or similar types of courses.

Exam categoryWeightInvigilationDurationGroupingComment exam
Exam category:
Submission
Form of assessment:
Written submission
Exam code:
GRA 41601
Grading scale:
Point scale leading to ECTS letter grade
Grading rules:
Internal and external examiner
Resit:
All components must, as a main rule, be retaken during next scheduled course
70No30 Hour(s)Individual
Exam category:
Activity
Form of assessment:
Presentation
Exam code:
GRA 41601
Grading scale:
Point scale leading to ECTS letter grade
Grading rules:
Internal and external examiner
Resit:
All components must, as a main rule, be retaken during next scheduled course
30No -Group ( 3 - 4)Presentation and defense of the mini project
Exams:
Exam category:Submission
Form of assessment:Written submission
Weight:70
Invigilation:No
Grouping (size):Individual
Duration:30 Hour(s)
Comment:
Exam code:GRA 41601
Grading scale:Point scale leading to ECTS letter grade
Resit:All components must, as a main rule, be retaken during next scheduled course
Exam category:Activity
Form of assessment:Presentation
Weight:30
Invigilation:No
Grouping (size):Group (3-4)
Duration: -
Comment:Presentation and defense of the mini project
Exam code:GRA 41601
Grading scale:Point scale leading to ECTS letter grade
Resit:All components must, as a main rule, be retaken during next scheduled course
Type of Assessment: 
Continuous assessment
Grading scale: 
ECTS
Total weight: 
100
Student workload
ActivityDurationComment
Teaching
36 Hour(s)
Group work / Assignments
100 Hour(s)
Prepare for teaching
12 Hour(s)
Examination
16 Hour(s)
Sum workload: 
164

A course of 1 ECTS credit corresponds to a workload of 26-30 hours. Therefore a course of 6 ECTS credits corresponds to a workload of at least 160 hours.