Sperimentazione e confronto di soluzioni per il rilevamento della mascherina sanitaria facciale

L’obiettivo principale della tesi, è quello di mettere a confronto soluzioni basate su tecnologie diverse e individuare la soluzione migliore che permetta di stabilire se le persone inquadrate in un’immagine indossano correttamente o meno la mascherina protettiva come previsto dalle norme anti-covid. Per raggiungere l’obiettivo verranno confrontate diverse architetture costruite per lo stesso scopo e che si basano sui principi di Machine Learning e Deep Learning, e verranno messe in funzione su insieme di dataset individuati, che sono stati creati per propositi affini.

Abstractive Long Document Summarization: Studio e Sperimentazione di Modelli Generativi Retrieval-Augmented

In questa tesi si trattano lo studio e la sperimentazione di un modello generativo retrieval-augmented, basato su Transformers, per il task di Abstractive Summarization su lunghe sentenze legali. La sintesi automatica del testo (Automatic Text Summarization) è diventata un task di Natural Language Processing (NLP) molto importante oggigiorno, visto il grandissimo numero di dati provenienti dal web e banche dati. Inoltre, essa permette di automatizzare un processo molto oneroso per gli esperti, specialmente nel settore legale, in cui i documenti sono lunghi e complicati, per cui difficili e dispendiosi da riassumere. I modelli allo stato dell’arte dell’Automatic Text Summarization sono basati su soluzioni di Deep Learning, in particolare sui Transformers, che rappresentano l’architettura più consolidata per task di NLP. Il modello proposto in questa tesi rappresenta una soluzione per la Long Document Summarization, ossia per generare riassunti di lunghe sequenze testuali. In particolare, l’architettura si basa sul modello RAG (Retrieval-Augmented Generation), recentemente introdotto dal team di ricerca Facebook AI per il task di Question Answering. L’obiettivo consiste nel modificare l’architettura RAG al fine di renderla adatta al task di Abstractive Long Document Summarization. In dettaglio, si vuole sfruttare e testare la memoria non parametrica del modello, con lo scopo di arricchire la rappresentazione del testo di input da riassumere. A tal fine, sono state sperimentate diverse configurazioni del modello su diverse tipologie di esperimenti e sono stati valutati i riassunti generati con diverse metriche automatiche.

Interpretable Convolutional Neural Networks for Decoding and Analyzing Neural Time Series Data

Machine learning is widely adopted to decode multi-variate neural time series, including electroencephalographic (EEG) and single-cell recordings. Recent solutions based on deep learning (DL) outperformed traditional decoders by automatically extracting relevant discriminative features from raw or minimally pre-processed signals. Convolutional Neural Networks (CNNs) have been successfully applied to EEG and are the most common DL-based EEG decoders in the state-of-the-art (SOA). However, the current research is affected by some limitations. SOA CNNs for EEG decoding usually exploit deep and heavy structures with the risk of overfitting small datasets, and architectures are often defined empirically. Furthermore, CNNs are mainly validated by designing within-subject decoders. Crucially, the automatically learned features mainly remain unexplored; conversely, interpreting these features may be of great value to use decoders also as analysis tools, highlighting neural signatures underlying the different decoded brain or behavioral states in a data-driven way. Lastly, SOA DL-based algorithms used to decode single-cell recordings rely on more complex, slower to train and less interpretable networks than CNNs, and the use of CNNs with these signals has not been investigated. This PhD research addresses the previous limitations, with reference to P300 and motor decoding from EEG, and motor decoding from single-neuron activity. CNNs were designed light, compact, and interpretable. Moreover, multiple training strategies were adopted, including transfer learning, which could reduce training times promoting the application of CNNs in practice. Furthermore, CNN-based EEG analyses were proposed to study neural features in the spatial, temporal and frequency domains, and proved to better highlight and enhance relevant neural features related to P300 and motor states than canonical EEG analyses. Remarkably, these analyses could be used, in perspective, to design novel EEG biomarkers for neurological or neurodevelopmental disorders. Lastly, CNNs were developed to decode single-neuron activity, providing a better compromise between performance and model complexity.

Big Data approaches as a support for precision livestock farming techniques

With the advent of new technologies it is increasingly easier to find data of different nature from even more accurate sensors that measure the most disparate physical quantities and with different methodologies. The collection of data thus becomes progressively important and takes the form of archiving, cataloging and online and offline consultation of information. Over time, the amount of data collected can become so relevant that it contains information that cannot be easily explored manually or with basic statistical techniques. The use of Big Data therefore becomes the object of more advanced investigation techniques, such as Machine Learning and Deep Learning. In this work some applications in the world of precision zootechnics and heat stress accused by dairy cows are described. Experimental Italian and German stables were involved for the training and testing of the Random Forest algorithm, obtaining a prediction of milk production depending on the microclimatic conditions of the previous days with satisfactory accuracy. Furthermore, in order to identify an objective method for identifying production drops, compared to the Wood model, typically used as an analytical model of the lactation curve, a Robust Statistics technique was used. Its application on some sample lactations and the results obtained allow us to be confident about the use of this method in the future.

Supervised and weakly supervised counting-by-segmentation: the fluorescent microscopy use case

This thesis focuses on automating the time-consuming task of manually counting activated neurons in fluorescent microscopy images, which is used to study the mechanisms underlying torpor. The traditional method of manual annotation can introduce bias and delay the outcome of experiments, so the author investigates a deep-learning-based procedure to automatize this task. The author explores two of the main convolutional-neural-network (CNNs) state-of-the-art architectures: UNet and ResUnet family model, and uses a counting-by-segmentation strategy to provide a justification of the objects considered during the counting process. The author also explores a weakly-supervised learning strategy that exploits only dot annotations. The author quantifies the advantages in terms of data reduction and counting performance boost obtainable with a transfer-learning approach and, specifically, a fine-tuning procedure. The author released the dataset used for the supervised use case and all the pre-training models, and designed a web application to share both the counting process pipeline developed in this work and the models pre-trained on the dataset analyzed in this work.

Data driven regularization models of non-linear ill-posed inverse problems in imaging

Imaging technologies are widely used in application fields such as natural sciences, engineering, medicine, and life sciences. A broad class of imaging problems reduces to solve ill-posed inverse problems (IPs). Traditional strategies to solve these ill-posed IPs rely on variational regularization methods, which are based on minimization of suitable energies, and make use of knowledge about the image formation model (forward operator) and prior knowledge on the solution, but lack in incorporating knowledge directly from data. On the other hand, the more recent learned approaches can easily learn the intricate statistics of images depending on a large set of data, but do not have a systematic method for incorporating prior knowledge about the image formation model. The main purpose of this thesis is to discuss data-driven image reconstruction methods which combine the benefits of these two different reconstruction strategies for the solution of highly nonlinear ill-posed inverse problems. Mathematical formulation and numerical approaches for image IPs, including linear as well as strongly nonlinear problems are described. More specifically we address the Electrical impedance Tomography (EIT) reconstruction problem by unrolling the regularized Gauss-Newton method and integrating the regularization learned by a data-adaptive neural network. Furthermore we investigate the solution of non-linear ill-posed IPs introducing a deep-PnP framework that integrates the graph convolutional denoiser into the proximal Gauss-Newton method with a practical application to the EIT, a recently introduced promising imaging technique. Efficient algorithms are then applied to the solution of the limited electrods problem in EIT, combining compressive sensing techniques and deep learning strategies. Finally, a transformer-based neural network architecture is adapted to restore the noisy solution of the Computed Tomography problem recovered using the filtered back-projection method.

Exploiting molecular simulations as an atomic-level microscope to elucidate clinically relevant cation chloride cotransporters

The cation chloride cotransporters (CCCs) represent a vital family of ion transporters, with several members implicated in significant neurological disorders. Specifically, conditions such as cerebrospinal fluid accumulation, epilepsy, Down’s syndrome, Asperger’s syndrome, and certain cancers have been attributed to various CCCs. This thesis delves into these pharmacological targets using advanced computational methodologies. I primarily employed GPU-accelerated all-atom molecular dynamics simulations, deep learning-based collective variables, enhanced sampling methods, and custom Python scripts for comprehensive simulation analyses. Our research predominantly centered on KCC1 and NKCC1 transporters. For KCC1, I examined its equilibrium dynamics in the presence/absence of an inhibitor and assessed the functional implications of different ion loading states. In contrast, our work on NKCC1 revealed its unique alternating access mechanism, termed the rocking-bundle mechanism. I identified a previously unobserved occluded state and demonstrated the transporter's potential for water permeability under specific conditions. Furthermore, I confirmed the actual water flow through its permeable states. In essence, this thesis leverages cutting-edge computational techniques to deepen our understanding of the CCCs, a family of ion transporters with profound clinical significance.

Quantitative susceptibility mapping as biomarker of neurodegeneration: methodological models and clinical applications

Quantitative Susceptibility Mapping (QSM) is an advanced magnetic resonance technique that can quantify in vivo biomarkers of pathology, such as alteration in iron and myelin concentration. It allows for the comparison of magnetic susceptibility properties within and between different subject groups. In this thesis, QSM acquisition and processing pipeline are discussed, together with clinical and methodological applications of QSM to neurodegeneration. In designing the studies, significant emphasis was placed on results reproducibility and interpretability. The first project focuses on the investigation of cortical regions in amyotrophic lateral sclerosis. By examining various histogram susceptibility properties, a pattern of increased iron content was revealed in patients with amyotrophic lateral sclerosis compared to controls and other neurodegenerative disorders. Moreover, there was a correlation between susceptibility and upper motor neuron impairment, particularly in patients experiencing rapid disease progression. Similarly, in the second application, QSM was used to examine cortical and sub-cortical areas in individuals with myotonic dystrophy type 1. The thalamus and brainstem were identified as structures of interest, with relevant correlations with clinical and laboratory data such as neurological evaluation and sleep records. In the third project, a robust pipeline for assessing radiomic susceptibility-based features reliability was implemented within a cohort of patients with multiple sclerosis and healthy controls. Lastly, a deep learning super-resolution model was applied to QSM images of healthy controls. The employed model demonstrated excellent generalization abilities and outperformed traditional up-sampling methods, without requiring a customized re-training. Across the three disorders investigated, it was evident that QSM is capable of distinguishing between patient groups and healthy controls while establishing correlations between imaging measurements and clinical data. These studies lay the foundation for future research, with the ultimate goal of achieving earlier and less invasive diagnoses of neurodegenerative disorders within the context of personalized medicine.

Addressing challenges in heterogeneous embedded systems programming: security and performance

Embedded systems are increasingly integral to daily life, improving and facilitating the efficiency of modern Cyber-Physical Systems which provide access to sensor data, and actuators. As modern architectures become increasingly complex and heterogeneous, their optimization becomes a challenging task. Additionally, ensuring platform security is important to avoid harm to individuals and assets. This study primarily addresses challenges in contemporary Embedded Systems, focusing on platform optimization and security enforcement. The initial section of this study delves into the application of machine learning methods to efficiently determine the optimal number of cores for a parallel RISC-V cluster to minimize energy consumption using static source code analysis. Results demonstrate that automated platform configuration is not only viable but also that there is a moderate performance trade-off when relying solely on static features. The second part focuses on addressing the problem of heterogeneous device mapping, which involves assigning tasks to the most suitable computational device in a heterogeneous platform for optimal runtime. The contribution of this section lies in the introduction of novel pre-processing techniques, along with a training framework called Siamese Networks, that enhances the classification performance of DeepLLVM, an advanced approach for task mapping. Importantly, these proposed approaches are independent from the specific deep-learning model used. Finally, this research work focuses on addressing issues concerning the binary exploitation of software running in modern Embedded Systems. It proposes an architecture to implement Control-Flow Integrity in embedded platforms with a Root-of-Trust, aiming to enhance security guarantees with limited hardware modifications. The approach involves enhancing the architecture of a modern RISC-V platform for autonomous vehicles by implementing a side-channel communication mechanism that relays control-flow changes executed by the process running on the host core to the Root-of-Trust. This approach has limited impact on performance and it is effective in enhancing the security of embedded platforms.

High-resolution stratigraphy and seismic site characterization of late Quaternary paleovalley systems (Adriatic coastal plain, Italy)

In highly urbanized coastal lowlands, effective site characterization is crucial for assessing seismic risk. It requires a comprehensive stratigraphic analysis of the shallow subsurface, coupled with the precise assessment of the geophysical properties of buried deposits. In this context, late Quaternary paleovalley systems, shallowly buried fluvial incisions formed during the Late Pleistocene sea-level fall and filled during the Holocene sea-level rise, are crucial for understanding seismic amplification due to their soft sediment infill and sharp lithologic contrasts. In this research, we conducted high-resolution stratigraphic analyses of two regions, the Pescara and Manfredonia areas along the Adriatic coastline of Italy, to delineate the geometries and facies architecture of two paleovalley systems. Furthermore, we carried out geophysical investigations to characterize the study areas and perform seismic response analyses. We tested the microtremor-based horizontal-to-vertical spectral ratio as a mapping tool to reconstruct the buried paleovalley geometries. We evaluated the relationship between geological and geophysical data and identified the stratigraphic surfaces responsible for the observed resonances. To perform seismic response analysis of the Pescara paleovalley system, we integrated the stratigraphic framework with microtremor and shear wave velocity measurements. The seismic response analysis highlights strong seismic amplifications in frequency ranges that can interact with a wide variety of building types. Additionally, we explored the applicability of artificial intelligence in performing facies analysis from borehole images. We used a robust dataset of high-resolution digital images from continuous sediment cores of Holocene age to outline a novel, deep-learning-based approach for performing automatic semantic segmentation directly on core images, leveraging the power of convolutional neural networks. We propose an automated model to rapidly characterize sediment cores, reproducing the sedimentologist's interpretation, and providing guidance for stratigraphic correlation and subsurface reconstructions.

studio e sperimentazione di tecniche di valutazione della significatività statistica in modelli nlp

Negli ultimi anni, il natural language processing ha subito una forte evoluzione, principalmente dettata dai paralleli avanzamenti nell’area del deep-learning. Con dimensioni architetturali in crescita esponenziale e corpora di addestramento sempre più comprensivi, i modelli neurali sono attualmente in grado di generare testo in maniera indistinguibile da quello umano. Tuttavia, a predizioni accurate su task complessi, si contrappongono metriche frequentemente arretrate, non capaci di cogliere le sfumature semantiche o le dimensioni di valutazione richieste. Tale divario motiva ancora oggi l’adozione di una valutazione umana come metodologia standard, ma la natura pervasiva del testo sul Web rende evidente il bisogno di sistemi automatici, scalabili, ed efficienti sia sul piano dei tempi che dei costi. In questa tesi si propone un’analisi delle principali metriche allo stato dell’arte per la valutazione di modelli pre-addestrati, partendo da quelle più popolari come Rouge fino ad arrivare a quelle che a loro volta sfruttano modelli per valutare il testo. Inoltre, si introduce una nuova libreria – denominata Blanche– finalizzata a raccogliere in un unico ambiente le implementazioni dei principali contributi oggi disponibili, agevolando il loro utilizzo da parte di sviluppatori e ricercatori. Infine, si applica Blanche per una valutazione ad ampio spettro dei risultati generativi ottenuti all’interno di un reale caso di studio, incentrato sulla verbalizzazione di eventi biomedici espressi nella letteratura scientifica. Una particolare attenzione è rivolta alla gestione dell’astrattività, un aspetto sempre più cruciale e sfidante sul piano valutativo.

Reti neurali "physics-informed" per modelli epidemiologici

La malattia COVID-19 associata alla sindrome respiratoria acuta grave da coronavirus 2 (SARS-CoV-2) ha rappresentato una grave minaccia per la salute pubblica e l’economia globale sin dalla sua scoperta in Cina, nel dicembre del 2019. Gli studiosi hanno effettuato numerosi studi ed in particolar modo l’applicazione di modelli epidemiologici costruiti a partire dai dati raccolti, ha permesso la previsione di diversi scenari sullo sviluppo della malattia, nel breve-medio termine. Gli obiettivi di questa tesi ruotano attorno a tre aspetti: i dati disponibili sulla malattia COVID-19, i modelli matematici compartimentali, con particolare riguardo al modello SEIJDHR che include le vaccinazioni, e l’utilizzo di reti neurali ”physics-informed” (PINNs), un nuovo approccio basato sul deep learning che mette insieme i primi due aspetti. I tre aspetti sono stati dapprima approfonditi singolarmente nei primi tre capitoli di questo lavoro e si sono poi applicate le PINNs al modello SEIJDHR. Infine, nel quarto capitolo vengono riportati frammenti rilevanti dei codici Python utilizzati e i risultati numerici ottenuti. In particolare vengono mostrati i grafici sulle previsioni nel breve-medio termine, ottenuti dando in input dati sul numero di positivi, ospedalizzati e deceduti giornalieri prima riguardanti la città di New York e poi l’Italia. Inoltre, nell’indagine della parte predittiva riguardante i dati italiani, si è individuato un punto critico legato alla funzione che modella la percentuale di ricoveri; sono stati quindi eseguiti numerosi esperimenti per il controllo di tali previsioni.

An Image Recognition Software for Identification of Vehicle Homologation Tags

In recent years, we have witnessed great changes in the industrial environment as a result of the innovations introduced by Industry 4.0, especially in the integration of Internet of Things, Automation and Robotics in the manufacturing field. The project presented in this thesis lies within this innovation context and describes the implementation of an Image Recognition application focused on the automotive field. The project aims at helping the supply chain operator to perform an effective and efficient check of the homologation tags present on vehicles. The user contribution consists in taking a picture of the tag and the application will automatically, exploiting Amazon Web Services, return the result of the control about the correctness of the tag, the correct positioning within the vehicle and the presence of faults or defects on the tag. To implement this application we ombined two IoT platforms widely used in industrial field: Amazon Web Services(AWS) and ThingWorx. AWS exploits Convolutional Neural Networks to perform Text Detection and Image Recognition, while PTC ThingWorx manages the user interface and the data manipulation.

Intermediate linguistic task fine-tuning on Multi-lingual models

State-of-the-art NLP systems are generally based on the assumption that the underlying models are provided with vast datasets to train on. However, especially when working in multi-lingual contexts, datasets are often scarce, thus more research should be carried out in this field. This thesis investigates the benefits of introducing an additional training step when fine-tuning NLP models, named Intermediate Training, which could be exploited to augment the data used for the training phase. The Intermediate Training step is applied by training models on NLP tasks that are not strictly related to the target task, aiming to verify if the models are able to leverage the learned knowledge of such tasks. Furthermore, in order to better analyze the synergies between different categories of NLP tasks, experimentations have been extended also to Multi-Task Training, in which the model is trained on multiple tasks at the same time.

Reti Generative Avversarie nel contesto di servizi di rete sociale: un caso di studio

L’intelligenza artificiale è senza dubbio uno degli argomenti attualmente più in voga nel mondo dell’informatica, sempre in costante evoluzione ed espansione in nuovi settori. In questa elaborato progettuale viene combinato l’argomento sopracitato con il mondo dei social network, che ormai sono parte integrante della quotidianità di tutti. Viene infatti analizzato lo stato dell’arte attuale delle reti neurali, in particolare delle reti generative avversarie, e vengono esaminate le principali tipologie di social network. Su questa base, infatti, verrà realizzato un sistema di rete sociale completo nel quale una GAN sarà proprio la protagonista, sfruttando le più interessanti tecnologie attualmente disponibili. Il sistema sarà disponibile sia come applicativo per dispositivi mobile che come sito web e introdurrà elementi di gamification per aumentare l’interazione con l’utente.