Machine Learning

Machine learning has emerged in the last decade as a powerful data analysis and prediction tool in many sectors, from Medical Imaging to Autonomous Driving. The study of the environment and of climate is no exception. Thanks to its ability to analyze and learn from vast amounts of data, it is an ideal instrument for many climate-related applications, such as predicting weather patterns, identifying sources of greenhouse gas emissions, optimizing energy systems, and monitoring wildlife populations, just to mention a few.

Through this course, students will learn about the basic machine-learning techniques for clustering, classification, and prediction. They will also learn how to preprocess and analyze large datasets, extract relevant features, and evaluate the performance of machine learning models.

This course introduces students to the application of machine learning techniques to climate change research. The course covers the basics of machine learning tools to analyze and model climate data from theoretical and practical perspectives. Students will learn how to apply machine learning algorithms to solve problems related to climate change, such as predicting future climate trends and understanding the impacts of climate change on natural and human systems.

The following arguments will be addressed:

1. Knowledge and understanding:
- understanding the theoretical bases of the main algorithms presented during lectures;
- understanding principles and differences of non-supervised learning algorithms;
- understanding principles and differences of supervised learning algorithms.

2. Applying knowledge and understanding in practical situations:
- being able to apply proper supervised and unsupervised analysis techniques to data;
- being able to use data analysis software tools used during lectures (e.g., scikit-learn);
- being able to compare and correctly interpret different analysis results from different algorithms

Basic knowledge of Python coding and Linear Algebra will be beneficial for a deeper understanding of both theoretical and practical aspects of Machine Learning.

1. Introduction to Data Science
- What is Machine Learning and Data Mining: concepts of supervised and unsupervised approaches
- Kinds of data
2. Clustering:
- Dimensionality reduction
- Clustering quality evaluation;
3. Supervised Learning
- Model training, validation and tuning; Feature Engineering
- Classification; Regression;
- Neural Networks (MLP and RNNs)

Students will be required to write a short paper on either:
- an (original) application of machine learning to a climate change related problem
- a detailed literature review of existing methods tackling a specific problem in climate change

scritto

Il/la docente ha il dovere di vigilare affinché siano rispettate le regole di autenticità e originalità delle prove d'esame. Di conseguenza, nei casi in cui vi sia il sospetto di un comportamento irregolare, l'esame può prevedere un ulteriore approfondimento, contestuale alla prova d'esame, che potrà essere realizzato anche in modalità differente rispetto alle modalità sopra riportate.

Il voto finale è calcolato come media pesata della prova scritta del primo modulo (50%), della prova scritta del secondo modulo (30%) e del progetto (20%).
Il voto finale viene assegnato secondo i seguenti criteri:
- 18–22: preparazione sufficiente, con conoscenze essenziali e applicazione parziale delle tecniche.
- 23–26: buona padronanza dei contenuti teorici e operativi, con corretto utilizzo degli strumenti richiesti.
- 27–30: preparazione completa, capacità di applicare in modo efficace le conoscenze a problemi complessi.
- 30 e lode: eccellenza in tutte le prove, autonomia e rigore nello sviluppo e nella presentazione del progetto.

Lectures and hands-on sessions.