ACIT4510 Statistical Learning Course description

The course covers the foundations and recent advances in Machine Learning from the point of view of Statistical Learning Theory. The goal of this course is to provide students with the practical skills to support the theoretical knowledge acquired during the lecture course and the practical intuitions needed to use and develop effective machine learning solutions to challenging problems.

Access to good statistical/data analysis software is paramount. Therefore, we will illustrate the use of the models throughout the course with real implementation.

No formal requirements over and above the admission requirements.

The student should have the following outcomes upon completing the course:

Knowledge

Upon successful completion of the course, the student:

• will have a good understanding the different concepts and methods of supervised and unsupervised statistical learning and how to apply them on large data.
• has advanced knowledge of probabilistic formulation of the various learning problems

;

Skills

Upon successful completion of the course, the student:

• can apply different high-dimensional regression techniques on data
• can apply different classification techniques on data
• can apply clustering techniques on data
• can derive learning algorithms for new models and analyze new data with them.
• can apply dimensionality reduction techniques on data

;

General competence

Upon successful completion of the course, the student:

• can apply different predictive models on data and assess their performance
• can use supervised and unsupervised learning in different real life problem

This course is divided into two parts. The first part with focus on covering the principles of Statistical Learning. Different seminars will be given on the different methodological aspects of Statistical learning, mainly, supervised learning and unsupervised learning.

The second part will focus on the students completing a programming project. This is a real data analysis problem, where the student is asked to carry out the analysis using the tools and techniques from the course and hand in a report documenting the steps he has taken in the analysis. The ultimate goal is to build a predictive model.

The project report will consist of at least 25 pages and max 60 pages.

During this part, there may be lectures if needed, but most of the time will be spent on individual supervision of students in lab-sessions.

Practical training

Lab sessions.

The following required coursework must be approved before the student can take the exam:

One mandatory assignment: A project plan document containing a description of the chosen data set, a preliminary research question and suggested tools and method to apply.

Signal processing defines the mathematical tools used to analyze, model and operate on measured data from physical signals and their sources. In digital signal processing (DSP) the following topics are lectured: the different domains for describing discrete signals and linear time invariant systems, digital filters and frequency analysis. Images are two-dimensional signals and similar;analytic methods apply.;

All aids are permitted.

No formal requirements over and above the admission requirements.

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

Knowledge

The course will give the students in depth knowledge on the following topics;

• Sampling, digitalization and;reconstruction.;
• How to obtain the signal spectrum of digital and analog;signals;
• How to obtain the frequency response of digital;systems;
• Methods of filtering discrete signals
• Digital image formats

;

Skills

The student will know how to:;

• Describe digital signals and systems mathematically in the time domain
• Describe discrete signals and systems in the frequency domain and the Z-domain;
• Describe digital images mathematically in real space and frequency space;
• Describe linear time invariant systems using difference equations, impulse responses,;and transfer functions;
• Analyze time discrete systems in the frequency domain ;
• Use discrete filters and the digitized versions of analog filters: FIR and;IIR;
• Apply post processing of images for filtering, noise reduction, edge detection and presentation.

;

General competence

The student will have general competence on:

• Frequency spectrums, impulse responses,;frequency responses,; convolution and;modulation;
• Fourier-series (FS), Fourier transform;(FT).;
• Sampling, reconstruction and;aliasing;
• Implementation of DSP-filters
• Improving image presentation

Lectures.

Exercises;

Computer exercises.