Tesi di dottorato

Tesi di dottorato: [2032] Home page macrotipologia

Sfoglia

Tesi di dottorato 2032

Prodotti recenti

Understanding Humans in Videos: From Pose to Action Recognition

This doctoral dissertation focuses on designing, developing, and evaluating methodologies for representing and understanding human motion and humans in the scene (interactions with objects or in small groups). The motivation arises from the growing interest in vision-based solutions which can understand and anticipate human behaviours, not only for robotics but also for surveillance or rehabilitation applications. Associated with specific objectives of this thesis, three distinct methodolo...

The Higgs, the Beauty and the Charm: Improving Jet Flavour-Tagging and Higgs Boson Measurements with Graph Neural Networks in the ATLAS Experiment at the LHC.

Testing the Yukawa couplings of the Higgs boson with fermions is essential to understanding the origin of fermion masses. Higgs boson decays to quark pairs are an important probe of these couplings, and of properties of the Higgs boson more generally. This thesis presents a novel measurement of the H -> bb and H -> cc decay rates in the VH production channel, where V stands for the vector bosons W and Z, using a dataset comprising 140 fb^−1 of proton-proton collision data collected by...

Large-scale convex optimization: parallelization and variance reduction

In this work, we investigate two aspects of large-scale optimization for convex functions defined on an infinite-dimensional separable Hilbert space: parallelized methods and incremental methods. These methods are used to efficiently solve problems that arise in data science, especially in machine learning and inverse problems. In parallelized optimization methods, the computational load of running the algorithm is distributed among several workers. For example, if the algorithm comprises a ...

Implicit Regularization: Insights from Iterative Regularization and Overparameterization

The goal of many data-driven problems is to achieve a good prediction by estimating a quantity of interest based on a finite set of (possibly noisy) measurements, exploiting training data, and some property of the model that may be known or not a-priori. The most common methods to reach this objective are explicit and implicit regularization. The first technique consists of minimizing the sum of a loss function plus a regularizer, which is explicitly added to the objective function and enta...

Interior Design of Cruise Ships: A Smart Materials-Driven Approach

Before defining the objectives and methodology of my doctoral research, it is important to understand the context in which this proposal was developed. We are in the middle of June 2020, three months after the achievement of the Master Degree in Nautical Engineering. In that situation, the results of a two-year project process were presented, related to a catamaran for passenger transport. This study allowed me to get ad advanced know-how related to the design spiral of a commercial vessel, b...

Scopri

Accesso al fulltext

open1883
no fulltext79
embargoed67
partially open2
reserved1