You are here:

99513 - ARTIFICIAL INTELLIGENCE AND MACHINE LEARNING FOR CLIMATE SCIENCE

Academic Year 2025/2026

                
                        Docente:
                        Nico Curti
                    
                        Credits:
                        6
                    
                        SSD:
                        FIS/07
                    
                        Language:
                        English
                    
                        Moduli:
                        
                            Michele Martinazzo
                            (Modulo 1)
                        
                            Giovanni De Cillis
                            (Modulo 2)
                        
                            Nico Curti
                            (Modulo 3)
                        
                        Teaching Mode:
                        
                                    In-person learning (entirely or partially) (Modulo 1); 
                                
                                    In-person learning (entirely or partially) (Modulo 2); 
                                
                                    In-person learning (entirely or partially) (Modulo 3)
                                
                            Campus:
                            Bologna
                        
                            Corso:
                            Second cycle degree programme (LM) in
                            Science of Climate (cod. 6697)

                            Teaching resources on Virtuale
                        
                                        Course Timetable
                                    
from Feb 24, 2026 to Jun 03, 2026

Learning outcomes

The student will learn the basis of artificial intelligence and machine learning (clustering, classification, regression, neural networks and Bayesian methods) and its application to climate science, with particular attention to the analysis of time series and space-resolved data. The student will learn how to implement these methods in open-source environments (eg. Python and R).

Course contents

Introduction to data analysis and machine learning fundamentals: features, data representation, supervised/unsupervised learning, regression, and classification
Introduction to climate data and climate science
Statistics: probability distributions, marginal and conditional probability, Bayesian theory
Machine learning: prediction problem, overfitting/underfitting concepts, hyperparameters, bias-variance decomposition, cross-validation, and proper data management in learning processes
Linear regression: MLE, Bayesian linear regression, introduction to parameter regularization
Linear classification: Discriminant analysis, logistic regression
Classification models: Support Vector Machine, Decision Tree, Random Forest
Introduction to neural networks: Simple Perceptron, activation functions, feed-forward neural network, error backpropagation, loss functions, parameter regularization, stochastic gradient descent, optimizers
Dimensionality reduction: optimal projection, SVD, PCA, t-SNE
Clustering: clustering Hierarchical, K-Means, Spectral Clustering, DBscan
Introduction to Autoencoders: sampling methods, expectation-maximization, VAE, GAN, Diffusion models
Convolutional Neural Networks: Applications and Main Architectures
Recursive Neural Networks: RNN Models, LSTM
Applications and practical examples through hands-on sessions and Python code

Readings/Bibliography

Bishop - Machine learning and pattern recognition

Ian Goodfellow and Yoshua Bengio and Aaron Courville - Deep Learning

Teaching methods

Frontal lectures with slides and blackboard

Assessment methods

Project report with original research analysis (with oral exam integration when necessary)

Students with learning disorders and\or temporary or permanent disabilities: please, contact the office responsible (https://site.unibo.it/studenti-con-disabilita-e-dsa/en/for-students) as soon as possible so that they can propose acceptable adjustments. The request for adaptation must be submitted in advance (15 days before the exam date) to the lecturer, who will assess the appropriateness of the adjustments, taking into account the teaching objectives.

Teaching tools

Lectures, slides and public repositories with codes used during the lectures as case study and examples

Office hours

See the website of Nico Curti

See the website of Michele Martinazzo

See the website of Giovanni De Cillis