Machine Learning for Gravitational-wave Astronomy

Durée

10h Th, 20h Pr

Nombre de crédits

Enseignant

Maxime Fays

Langue(s) de l'unité d'enseignement

Langue anglaise

Organisation et évaluation

Enseignement au deuxième quadrimestre

Horaire

Horaire en ligne

Unités d'enseignement prérequises et corequises

Les unités prérequises ou corequises sont présentées au sein de chaque programme

Contenus de l'unité d'enseignement

This course introduces advanced machine learning techniques applied to gravitational-wave astronomy. Students will learn to identify and analyze gravitational-wave signals in noisy environments using machine learning algorithms.

The course will cover the following topics, which are tentative and subject to change based on emerging research:

• Gravitational-Wave Detection: Methods to detect gravitational waves from various astrophysical sources, including binary black holes and neutron stars.
• Signal Processing and Denoising: Techniques such as Sparse Dictionary Learning and Autoencoders to denoise gravitational wave signals and remove glitches from interferometer data.
• Time-Series Analysis: Long Short-Term Memory Networks (LSTMs) for processing time-series data, focusing on detecting transient gravitational-wave signals.
• Convolutional Neural Networks (CNNs): Applications of CNNs to recognize patterns in gravitational wave data, including time-frequency maps for signal classification and identification.
• Anomaly Detection: Using Convolutional Autoencoders for detecting anomalies in data, such as noise and glitches that impact the quality of gravitational-wave detections.
• Generative Adversarial Networks (GANs): Exploring the use of GANs to generate synthetic data for training and testing machine learning models, particularly for noise generation and data augmentation.
• XGBoost: Applying this powerful gradient boosting technique for classification and regression tasks in gravitational wave data analysis.

Acquis d'apprentissage (objectifs d'apprentissage) de l'unité d'enseignement

By the end of the course, students will be able to:

• Understand and apply advanced machine learning techniques to real gravitational-wave data.
• Analyze and classify noisy datasets from gravitational wave interferometers.
• Develop machine learning models for detecting, denoising, and classifying gravitational wave signals.
• Implement neural networks and boosting algorithms to solve complex challenges in gravitational wave astronomy.

Savoirs et compétences prérequis

Students should have a solid foundation in Python programming and basic machine learning concepts. Prior knowledge of gravitational-wave astronomy is also recommended.

Activités d'apprentissage prévues et méthodes d'enseignement

The course will combine lectures with practical hands-on projects. Students will work on real gravitational-wave datasets using Jupyter notebooks. Group discussions and student presentations will also be included to encourage collaborative learning.

Mode d'enseignement (présentiel, à distance, hybride)

Cours donné exclusivement en présentiel


Informations complémentaires:

The course will be delivered in person, with flexibility for online discussions or assignments as needed.

Supports de cours, lectures obligatoires ou recommandées

Plate-forme(s) utilisée(s) pour les supports de cours :
- MyULiège


Informations complémentaires:

Course materials will include selected readings from recent research papers on machine learning applied to gravitational-wave data, as well as Python tutorials for implementing the machine learning models.

Modalités d'évaluation et critères

Examen(s) en session

Toutes sessions confondues

- En présentiel

évaluation orale

Travail à rendre - rapport


Informations complémentaires:

• An oral exam to assess students' understanding of both gravitational-wave concepts and the machine learning techniques discussed during the course.
• A homework assignment that will be graded, focused on applying machine learning methods to gravitational-wave data.

Stage(s)

Remarques organisationnelles et modifications principales apportées au cours

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  

Association d'un ou plusieurs MOOCs