701-3003-00L Environmental Systems Data Science: Machine Learning
| Semester | Autumn Semester 2024 |
| Lecturers | L. Pellissier, C. P. Albouy, M. Volpi |
| Periodicity | yearly recurring course |
| Language of instruction | English |
| Abstract | Students are introduced to advanced data science where environmental data are analyzed using state of the art machine learning methods. Starting from known statistical approaches, they learn the principle of more advanced machine learning methods with practical application. The course enables students to plan their own data science project in their specialization and to apply machine learning mode |
| Learning objective | The students are able to • select an appropriate model related to a research question and dataset • describe the steps from data preparation to running and evaluating models • prepare data for running machine learning with dependent and independent variable • build and validate regressions and neural network models • understand convolution and deep learning models • access online resources to keep up with the latest data science methodology and deepen their understanding |
| Content | • The data science workflow • Data preparation for running and validating machine learning models • Get to know machine learning approaches including regression, random forest and neural network • Model complexity and hyperparameters • Model parameterization and loss • Model evaluations and uncertainty • Deep learning with convolutions |
| Literature | Building on existing data science resources |
| Prerequisites / Notice | Math IV, VI (Statistics); R, Python; ESDS I |