Objectives of Programme's Learning Outcomes

• Understand the fundamentals of computer science
• Show an understanding and deep knowledge of the concepts of computer science and mathematical formalisms used in the field of computer science
• Solve exercises and computer problems by applying basic knowledge in the various disciplines of computer science
• Use the vocabulary and the correct mathematical reasoning to formulate and solve problems in the field of computer science
• Use and combine knowledge from different disciplines to solve multidisciplinary problems
• Understand computer technologies
• Self-train in ICT
• Demonstrate basic knowledge and know-how in related fields
• Have a good knowledge of English in order to read and understand scientific texts, especially in the field of computer science.
• Demonstrate knowledge and basic skills in science and technology.
• Manage IT projects
• Manage a project in compliance with specifications, constraints and deadlines
• Creatively implement knowledge and expertise gained in the field of computer science.
• Apply appropriate technological and scientific ICT approaches
• Demonstrate independence and their ability to work in teams.
• Understand the fundamentals related to scientific methods
• Develop skills of abstraction and modelling through a conceptual and scientific approach
• Conduct rigorous reasoning based on scientific arguments
• Understand the fundamentals of communication
• Communicate information (both orally and in writing) relating to the field of computer science in an intelligible, clear and structured way
• Communicate a consistent and rigorous scientific argument, either orally or in writing
• Have a good command of language and communication techniques.

Learning Outcomes of UE

This course provides an introduction to the foundations of machine learning. It covers supervised learning (regression, classification) and unsupervised learning (dimensionality reduction), linear and nonlinear models (decision trees, neural networks), as well as ensemble methods (bagging, random forests, boosting). Particular attention is given to core concepts in statistical learning, such as the bias-variance trade-off, overfitting, model selection, and resampling techniques. The theoretical content is complemented by practical sessions in Python using the scikit-learn and PyTorch libraries.

UE Content: description and pedagogical relevance

See the single learning activity (AA).
 

Prior Experience

Probability and statistics (a few reminders are given at the start of the semester).
Basics of linear algebra.
Basics of non-linear optimization.