Advanced Applied Machine Learning
Avancerad tillämpad maskininlärning

EDA090F, 7.5 credits

Valid from: Spring 2023
Decided by: Maria Sandsten
Date of establishment: 2023-01-24

General Information

Division: Computer Science (LTH)
Course type: Third-cycle course
Teaching language: English

Aim

To give deeper understanding and skills within areas of machine learning that were introduced in the course EDAN95F / EDAN96 Applied Machine Learning, and to introduced new, advanced, topics that have not been discussed in the course EDAN95F / EDAN96.

Goals

Knowledge and Understanding

For a passing grade the doctoral student must

• - show knowledge and understanding within the different topics discussed in the course
• - be able to explain the how methods and models that form the course's core in terms of how they are constructed and how they work
• - be able to describe other methods and topics that have been discussed in the course on overview level

Competences and Skills

For a passing grade the doctoral student must

• - be able to handle software tools (introduced in the course) for data handling and pre-processing
• - be able to handle tools (introduced in the course) for and perform suitable analysis of methods and models applied in the course
• - be able to implement their own models / solutions within the course's core topics

Judgement and Approach

For a passing grade the doctoral student must

• - show understanding for which methods (machine learning models and algorithms) are suitable in which areas of applications / for which type of problem
• - show understanding of the limitations that exist in one way or the other for all machine learning approaches, as well as be able to describe these limitations and respective consequences for the application of the models and methods

Course Contents

The core content of the course consists of four topics: Image classification with deep neural networks (mainly CNNs), text processing with deep recurrent networks and transformers (e.g., RNNs), Bayesian optimisation, and Reinforcement Learning. Other topics will be discussed on overview level. The core topics are handled both theoretically and, through respective assignments, practically.

Course Literature

A literature listing is given in the course plan for EDAP30 Advanced Applied Machine Learning

Instruction Details

Types of instruction: Lectures, laboratory exercises. "Laboratory exercises" are programming assignments that can be worked on in small groups and that have to be presented to a TA in respective lab sessions or within a seminar session (entire group of graduate students in the course) led by a teacher

Examination Details

Examination formats: Written assignments, seminars given by participants. The laboratory sessions (see above) can be replaced with a seminar session for all graduate students on the course, in which the assignments are discussed with a teacher
Grading scale: Failed, pass
Examiner: Head of department Elin A. Topp

Admission Details

Assumed prior knowledge: EDAN95F / EDAN96 Applied Machine Learning, FMAN45 Machine Learning or comparable background

Course Occasion Information

Start date: 2023-03-20
Course pace: Full time

Application Information

Contact the course responsible teacher (Elin A. Topp) personally by e-mail. Places are assigned based on an evaluation of the situation in the course EDAP30, which this course is based on. Priority is given according to the following order of affiliations: Dept of Computer Science, LTH, LU, external.

Contact and Other Information

Course coordinators:

• Erik Hellsten <erik.hellsten@cs.lth.se>
• Luigi Nardi <luigi.nardi@cs.lth.se>