Machine Learning for engineers (ML4E) Initial proposal (to be fine-tuned) - 5x 3 hours lectures Contents / brief syllabus: + 60% : Machine Learning (ML) concepts and techniques - intro (past, present, future, ethics, …) - bias/variance trade-off - regression, classification, clustering, dimensionality reduction - basics of linear regression, random forests (RF), artificial neural networks (ANN), support vector machines (SVM), k-nearest neighbors (KNN), … + 40% : data-efficient Bayesian ML techniques, with focus on engineering design - Bayesian ML techniques, Gaussian Processes (GP) - Sequential experimental design, Active Learning (AL) - Bayesian optimization - Engineering applications, with focus on data-efficiency: examples from electronics, mechanics and fluid dynamics

-

This course provides the doctoral candidates with the opportunity to develop a broad understanding of machine learning. After learning about the prominent machine learning models and how to interpret metrics, the attendees will learn how to select the most appropriate model for an identified technical problem.

A distanza in modalità sincrona

Presentazione orale

P.D.1-1 - Febbraio