Duration
30h Th, 15h Pr
Number of credits
| Master of Science (MSc) in Data Science | 5 crédits | |||
| Master of Science (MSc) in Data Science and Engineering | 5 crédits |
Lecturer
Language(s) of instruction
English language
Organisation and examination
Teaching in the first semester, review in January
Schedule
Units courses prerequisite and corequisite
Prerequisite or corequisite units are presented within each program
Learning unit contents
This master-level course is designed to provide students with advanced knowledge and skills in applying machine learning techniques to space science research. Students will explore various aspects of space science, from satellite data analysis to astrophysical simulations, and learn how machine learning can enhance our understanding of the cosmos.
Through a combination of lectures, hands-on projects, and research assignments, participants will develop expertise in both space science and machine learning, making them well-equipped for careers in academia, research institutions, and space industry.
Space science topics to be covered (tentative and subject to change):
* Exoplanet Detection and Characterization
* Cosmic Microwave Background Analysis
* Gravitational Wave Detection and Analysis
* Stellar Evolution and Classification
* Galaxy Morphology and Classification
* Dark Matter and Dark Energy Studies
* Black Hole Imaging and Event Horizon Studies
* Solar Activity and Space Weather Prediction
* Astronomical Surveys and Sky Mapping
* Astroparticle Physics and Neutrino Detection
Machine learning techniques to be covered (tentative and subject to change):
* Random Forests
* Convolutional Neural Networks (CNNs)
* Long Short-Term Memory (LSTM) Networks
* Principal Component Analysis (PCA)
* Support Vector Machines (SVM)
* Recurrent Neural Networks (RNNs)
* K-Means Clustering
* Gaussian Mixture Models (GMMs)
* Transfer Learning with Pre-trained Models
* Generative Adversarial Networks (GANs)
Learning outcomes of the learning unit
Participants will develop expertise in both space science and machine learning, making them well-equipped for careers in academia, research institutions, and the space industry. Graduates of this course will be well positioned to develop innovative space-related algorithms, and drive the next generation of space science discoveries through data-driven approaches. Their interdisciplinary skills will be highly sought after in a rapidly evolving space science landscape.
Prerequisite knowledge and skills
///
Planned learning activities and teaching methods
This course is based on lectures, and on discussion sessions where problems are discussed & solved in the class. The problems will be solved by the students, under the guidance of the instructor.
Mode of delivery (face to face, distance learning, hybrid learning)
Face-to-face course
Additional information:
Face-to-face if possible, depending on the COVID situation.
Recommended or required readings
Statistics, Data Mining, and Machine Learning for Astronomy, Ivezic et al.
Exam(s) in session
Any session
- In-person
oral exam
Written work / report
Additional information:
The exam will consist of several questions concerning the theory as seen during the lectures.
Regarding the Machine Learning application, a take-home exam on a subject discussed together with the lecturer will be assigned to assess students' comprehension and practical application of machine learning techniques in space science.
Work placement(s)
Organisational remarks and main changes to the course
Each lesson will start with a short (%7E30 minutes, depending on the topic) theoretical introduction to the astrophysical concept necessary to establish a strong foundational understanding before delving into the practical aspects of applying machine learning techniques in space science, through the use of a jupyter notebook given by the lecturer.
Contacts
Maxime Fays
(maxime.fays@uliege.be)
Room 4.43 Bât. B5A
Inter. fondamentales en physique et astrophysique (IFPA)
Quartier Agora allée du six Août 19
4000 Liège
Téléphone de service: +32 4 3663643