Content, progress and pedagogy of the module

The module is based on knowledge in Linear Algebra, Calculus, and Probability Theory. Basic Python programming experience is recommended but not required.

Learning objectives

Knowledge

• Have knowledge about basic machine learning concepts including supervised / unsupervised learning, regression / classification tasks, model selection and validation
• Have knowledge about different neural network architectures (MLP, RNN, CNN, Generative networks etc.), including their main applications, strengths and weaknesses
• Have knowledge about basic data preprocessing techniques including normalisation, resampling, frequency extraction, etc.

Skills

• Be able to train and evaluate the performance of various deep learning models
• Be able to optimise the training of deep learning models through regularisation and hyperparameter tuning
• Be able to implement the models and methods using popular Python machine learning libraries, such as NumPy, Scikit-learn and PyTorch

Competences

• Be able to frame a concrete engineering problem into a machine learning framework including considerations of the task and available data
• Be able to identify relevant models, data processing methods and optimisation methods for a given engineering problem
• Be able to document and discuss the chosen model and methods and resulting performance

Type of instruction

Mixture of lectures, practical examples and exercises.

Extent and expected workload

Since it is a 5 ECTS course module, the work load is expected to be 150 hours for the student.

Exam

Exams