Standard Quadratic Optimization problems are an important family of problems with a large number of applications in finance, decision science, graphs algorithms, etc.For these nonconvex problems, we are interested in finding global optima; however, exact global optimization algorithms hardly scale
Advanced memetic algorithms for clustering
This topic is related to a paper we recently published on this subject: P. Mansueto and F. Schoen, “Memetic differential evolution methods for clustering problems“, Pattern Recognition, 114, 2021. The idea is to extend those methods and experiment with different
Exact optimization models for airport personnel scheduling
Student: Giulia Pellegrini (December 2021) In this thesis we explored the power and the limits of exact optimization models for large personnel scheduling problems. Scheduling at airports has the additional complication, with respect to other personnel scheduling problems, that shifts
Machine learning techniques for attributes extraction in apparel images
Student: Francesca Del Lungo, 2021. This thesis was developed during a stage at Intuendi srl, Florence Clothing classification algorithms often face several challenges. First of all,clothing items often have many variations in style, texture and cutting. Second, clothing attributes are
Optimization methods for risk scores learning
Student: Luca Ciabini, 2021 Risk scores are classi fication models that predict the risk of an event using a value de fined by the sum of a few integers. They are particularly interesting as they are easy to learn, to use, to
Machine learning models for sports betting on professional Tennis
Student: Lorenzo Amato, graduated: 2021
Image Captioning road scenes: a multitask deep learning approach
Student: Niccolò Bellaccini, 2020. Thesis developed partially during a stage at Verizon Connect Research Italy, Florence Image Captioning tackles the problem of generating textual descriptionsfrom pictures, and thus lies in the intersection of the Computer Vision (CV)and Natural Language Processing
Efficient algorithms based on optimal decision trees for transparent machine learning
Student: Tommaso Aldinucci In recent years the world of machine learning has been revolutionized by the “explosion” of models based on deep neural networks. Such models have proved to be really performing and able to face even very complex tasks.
A Lightweight Deep Machine Learning Model for Vehicle View-point Estimation from Dash-cam Images
This thesis was partially developed at Verizon Connect Research Italy, in Florence. Student: Simone Magistri. Vehicle pose estimation from vehicle cameras is a crucial component of road scene understanding. In this thesis we propose a deep light-weight method to predict
Advanced optimization methods for clustering
K-means is one out of a bunch of widely used clustering techniques. In this thesis, we plan to experiment with global optimization algorithms which use K-means as a local optimization solver Candidate: Pierluigi Mansueto From this thesis a paper was