Virtual Learning Factory on VR-Supported Factory Planning

Part 13: Virtual Reality and Simulation

Magnetic and Acoustic Investigations of Turbulent Spherical Couette Flow

Title of dissertation: MAGNETIC AND ACOUSTIC INVESTIGATIONS OF TURBULENT SPHERICAL COUETTE FLOW Matthew M. Adams, Doctor of Philosophy, 2016 Dissertation directed by: Professor Daniel Lathrop Department of Physics This dissertation describes experiments in spherical Couette devices, using both gas and liquid sodium. The experimental geometry is motivated by the Earth's outer core, the seat of the geodynamo, and consists of an outer spherical shell and an inner sphere, both of which can be rotated independently to drive a shear flow in the fluid lying between them. In the case of experiments with liquid sodium, we apply DC axial magnetic fields, with a dominant dipole or quadrupole component, to the system. We measure the magnetic field induced by the flow of liquid sodium using an external array of Hall effect magnetic field probes, as well as two probes inserted into the fluid volume. This gives information about possible velocity patterns present, and we extend previous work categorizing flow states, noting further information that can be extracted from the induced field measurements. The limitations due to a lack of direct velocity measurements prompted us to work on developing the technique of using acoustic modes to measure zonal flows. Using gas as the working fluid in our 60~cm diameter spherical Couette experiment, we identified acoustic modes of the container, and obtained excellent agreement with theoretical predictions. For the case of uniform rotation of the system, we compared the acoustic mode frequency splittings with theoretical predictions for solid body flow, and obtained excellent agreement. This gave us confidence in extending this work to the case of differential rotation, with a turbulent flow state. Using the measured splittings for this case, our colleagues performed an inversion to infer the pattern of zonal velocities within the flow, the first such inversion in a rotating laboratory experiment. This technique holds promise for use in liquid sodium experiments, for which zonal flow measurements have historically been challenging.

Zig-zagging in geometrical reasoning in technological collaborative environments: a Mathematical Working Space-framed study concerning cognition and affect

This study highlights the importance of cognition-affect interaction pathways in the construction of mathematical knowledge. Scientific output demands further research on the conceptual structure underlying such interaction aimed at coping with the high complexity of its interpretation. The paper discusses the effectiveness of using a dynamic model such as that outlined in the Mathematical Working Spaces (MWS) framework, in order to describe the interplay between cognition and affect in the transitions from instrumental to discursive geneses in geometrical reasoning. The results based on empirical data from a teaching experiment at a middle school show that the use of dynamic geometry software favours students’ attitudinal and volitional dimensions and helps them to maintain productive affective pathways, affording greater intellectual independence in mathematical work and interaction with the context that impact learning opportunities in geometric proofs. The reflective and heuristic dimensions of teacher mediation in students’ learning is crucial in the transition from instrumental to discursive genesis and working stability in the Instrumental-Discursive plane of MWS.

Tourism Destination Management: A Collaborative Approach

Part 6: Engineering and Implementation of Collaborative Networks

Synchronization of Heteroclinic Circuits Through Learning in Chains of Neural Motifs

The synchronization of oscillatory activity in networks of neural networks is usually implemented through coupling the state variables describing neuronal dynamics. In this study we discuss another but complementary mechanism based on a learning process with memory. A driver network motif, acting as a teacher, exhibits winner-less competition (WLC) dynamics, while a driven motif, a learner, tunes its internal couplings according to the oscillations observed in the teacher. We show that under appropriate training the learner motif can dynamically copy the coupling pattern of the teacher and thus synchronize oscillations with the teacher. Then, we demonstrate that the replication of the WLC dynamics occurs for intermediate memory lengths only. In a unidirectional chain of N motifs coupled through teacher-learner paradigm the time interval required for pattern replication grows linearly with the chain size, hence the learning process does not blow up and at the end we observe phase synchronized oscillations along the chain. We also show that in a learning chain closed into a ring the network motifs come to a consensus, i.e. to a state with the same connectivity pattern corresponding to the mean initial pattern averaged over all network motifs.

Phylogeography of the Atlanto-Mediterranean sea cucumber Holothuria (Holothuria) mammata: the combined effects of historical processes and current oceanographical pattern

We assessed the genetic structure of populations of the widely distributed sea cucumber Holothuria (Holothuria) mammata Grube, 1840, and investigated the effects of marine barriers to gene flow and historical processes. Several potential genetic breaks were considered, which would separate the Atlantic and Mediterranean basins, the isolated Macaronesian Islands from the other locations analysed, and the Western Mediterranean and Aegean Sea (Eastern Mediterranean). We analysed mitochondrial 16S and COI gene sequences from 177 individuals from four Atlantic locations and four Mediterranean locations. Haplotype diversity was high (H = 0.9307 for 16S and 0.9203 for COI), and the haplotypes were closely related (p = 0.0058 for 16S and 0.0071 for COI). The lowest genetic diversities were found in the Aegean Sea population. Our results showed that the COI gene was more variable and more useful for the detection of population structure than the 16S gene. The distribution of mtDNA haplotypes, the pairwise FST values and the results of exact tests and AMOVA revealed: (i) a significant genetic break between the population in the Aegean Sea and those in the other locations, as supported by both mitochondrial genes, and (ii) weak differentiation of the Canary and Azores Islands from the other populations; however, the populations from the Macaronesian Islands, Algarve and West Mediterranean could be considered to be a panmictic metapopulation. Isolation by distance was not identified in H. (H.) mammata. Historical events behind the observed findings, together with the current oceanographic patterns, were proposed and discussed as the main factors that determine the population structure and genetic signature of H. (H.) mammata

Project RAILS: Lessons Learned about Collaborative Rubric Assessment of Information Literacy Skills

Link to article on publisher site: https://www.press.jhu.edu/journals/portal_libraries_and_the_academy/portal_pre_print/articles/belanger.pdf

CrashEd – A live immersive, learning experience embedding STEM subjects in a realistic, interactive crime scene

Interactive experiences are rapidly becoming popular via the surge of ‘escape rooms’; part game and part theatre, the ‘escape’ experience is exploding globally, having gone from zero offered at the outset of 2010 to at least 2800 different experiences available worldwide today. CrashEd is an interactive learning experience that parallels many of the attractions of an escape room – it incorporates a staged, realistic ‘crime scene’ and invites participants to work together to gather forensic evidence and question a witness in order to solve a crime, all whilst competing against a ticking clock. An animation can enhance reality and engage with cognitive processes to help learning; in CrashEd, it is the last piece of the jigsaw that consolidates the students’ incremental acquisition of knowledge to tie together the pieces of evidence, identify a suspect and ultimately solve the crime. This article presents the background to CrashEd and an overview of how a timely placed animation at the end of an educational experience can enhance learning. The lessons learned, from delivering bespoke versions of the experience to different demographic groups, are discussed. The article will consider the successes and challenges raised by the collaborative project, future developments and potential wider implications of the development of CrashEd.

A take on complexity: bio-molecules and human metabolism interaction modelling for health and nutrition with machine learning

The advent of omic data production has opened many new perspectives in the quest for modelling complexity in biophysical systems. With the capability of characterizing a complex organism through the patterns of its molecular states, observed at different levels through various omics, a new paradigm of investigation is arising. In this thesis, we investigate the links between perturbations of the human organism, described as the ensemble of crosstalk of its molecular states, and health. Machine learning plays a key role within this picture, both in omic data analysis and model building. We propose and discuss different frameworks developed by the author using machine learning for data reduction, integration, projection on latent features, pattern analysis, classification and clustering of omic data, with a focus on 1H NMR metabolomic spectral data. The aim is to link different levels of omic observations of molecular states, from nanoscale to macroscale, to study perturbations such as diseases and diet interpreted as changes in molecular patterns. The first part of this work focuses on the fingerprinting of diseases, linking cellular and systemic metabolomics with genomic to asses and predict the downstream of perturbations all the way down to the enzymatic network. The second part is a set of frameworks and models, developed with 1H NMR metabolomic at its core, to study the exposure of the human organism to diet and food intake in its full complexity, from epidemiological data analysis to molecular characterization of food structure.

A machine learning based method to detect genomic imbalances exploiting X chromosome exome reads

Whole Exome Sequencing (WES) is rapidly becoming the first-tier test in clinics, both thanks to its declining costs and the development of new platforms that help clinicians in the analysis and interpretation of SNV and InDels. However, we still know very little on how CNV detection could increase WES diagnostic yield. A plethora of exome CNV callers have been published over the years, all showing good performances towards specific CNV classes and sizes, suggesting that the combination of multiple tools is needed to obtain an overall good detection performance. Here we present TrainX, a ML-based method for calling heterozygous CNVs in WES data using EXCAVATOR2 Normalized Read Counts. We select males and females’ non pseudo-autosomal chromosome X alignments to construct our dataset and train our model, make predictions on autosomes target regions and use HMM to call CNVs. We compared TrainX against a set of CNV tools differing for the detection method (GATK4 gCNV, ExomeDepth, DECoN, CNVkit and EXCAVATOR2) and found that our algorithm outperformed them in terms of stability, as we identified both deletions and duplications with good scores (0.87 and 0.82 F1-scores respectively) and for sizes reaching the minimum resolution of 2 target regions. We also evaluated the method robustness using a set of WES and SNP array data (n=251), part of the Italian cohort of Epi25 collaborative, and were able to retrieve all clinical CNVs previously identified by the SNP array. TrainX showed good accuracy in detecting heterozygous CNVs of different sizes, making it a promising tool to use in a diagnostic setting.

Sinus rhythm restoration with electrical cardioversion: acute effect of shock configuration and subsequent modifications in peripheral flow and sleep.

Atrial fibrillation (AF) is a widespread arrhythmia, associated with higher risk of stroke, sleep disorders and dementia. In some conditions, electrical cardioversion (ECV) represents the best choice for rhythm control. Nowadays, there is a growing interest in developing new devices for screening and monitoring of AF patients. We aimed to improve acute efficacy of ECV procedure and to explore the feasibility of the use of new wearable devices for monitoring in candidates to AF ECV. We compared antero-apical pads vs antero-posterior patches approach for AF ECV, and we elaborated a decision algorithm to improve acute efficacy. After, we evaluated the feasibility of the use of new wearable devices for monitoring of candidates to AF ECV. In particular, we analysed the effect of AF ECV on heart rate variability and vascular age parameters derived from PPG signals registered with Empatica (CE 1876/MDD 93/42/EEC), and on EEG pattern registered with Neurosteer (Israel). From December 2005 to September 2019, 492 patients were enrolled. We evaluated acute efficacy of the two approaches for AF ECV and we elaborated a decision algorithm based on body surface area, weight, and height. The decision algorithm improved first shock efficacy (93.2% vs. 87.2%, p=0.025). From 1st November 2021 to 1st April 2022, 24 patients were enrolled in PPEEG-AF pilot study. Considering vascular age parameters, a significant reduction in TPR and a wave was observed (p<0.001). Considering sleep patterns, a tendency to higher coherence was observed in registrations acquired during AF, or considering signals registered for each patient independently from AF. The new decision algorithm improved acute efficacy and reduced costs associated with adhesive patches. Significant modifications were observed on vascular age parameters measured before and after ECV, and a possible AF effect on sleep pattern was noticed. More data are necessary to confirm these preliminary results.

Integrated communication and learning towards autonomous mobile radio networks

The integration of distributed and ubiquitous intelligence has emerged over the last years as the mainspring of transformative advancements in mobile radio networks. As we approach the era of “mobile for intelligence”, next-generation wireless networks are poised to undergo significant and profound changes. Notably, the overarching challenge that lies ahead is the development and implementation of integrated communication and learning mechanisms that will enable the realization of autonomous mobile radio networks. The ultimate pursuit of eliminating human-in-the-loop constitutes an ambitious challenge, necessitating a meticulous delineation of the fundamental characteristics that artificial intelligence (AI) should possess to effectively achieve this objective. This challenge represents a paradigm shift in the design, deployment, and operation of wireless networks, where conventional, static configurations give way to dynamic, adaptive, and AI-native systems capable of self-optimization, self-sustainment, and learning. This thesis aims to provide a comprehensive exploration of the fundamental principles and practical approaches required to create autonomous mobile radio networks that seamlessly integrate communication and learning components. The first chapter of this thesis introduces the notion of Predictive Quality of Service (PQoS) and adaptive optimization and expands upon the challenge to achieve adaptable, reliable, and robust network performance in dynamic and ever-changing environments. The subsequent chapter delves into the revolutionary role of generative AI in shaping next-generation autonomous networks. This chapter emphasizes achieving trustworthy uncertainty-aware generation processes with the use of approximate Bayesian methods and aims to show how generative AI can improve generalization while reducing data communication costs. Finally, the thesis embarks on the topic of distributed learning over wireless networks. Distributed learning and its declinations, including multi-agent reinforcement learning systems and federated learning, have the potential to meet the scalability demands of modern data-driven applications, enabling efficient and collaborative model training across dynamic scenarios while ensuring data privacy and reducing communication overhead.

Machine learning techniques for heavy-flavour baryon production measurements at the LHC

Il quark-gluon plasma (QGP) è uno stato della materia previsto dalla cromodinamica quantistica. L’esperimento ALICE a LHC ha tra i suoi obbiettivi principali lo studio della materia fortemente interagente e le proprietà del QGP attraverso collisioni di ioni pesanti ultra-relativistici. Per un’esaustiva comprensione di tali proprietà, le stesse misure effettuate su sistemi collidenti più piccoli (collisioni protone-protone e protone-ione) sono necessarie come riferimento. Le recenti analisi dei dati raccolti ad ALICE hanno mostrato che la nostra comprensione dei meccanismi di adronizzazione di quark pesanti non è completa, perchè i dati ottenuti in collisioni pp e p-Pb non sono riproducibili utilizzando modelli basati sui risultati ottenuti con collisioni e+e− ed ep. Per questo motivo, nuovi modelli teorici e fenomenologici, in grado di riprodurre le misure sperimentali, sono stati proposti. Gli errori associati a queste nuove misure sperimentali al momento non permettono di verificare in maniera chiara la veridicità dei diversi modelli proposti. Nei prossimi anni sarà quindi fondamentale aumentare la precisione di tali misure sperimentali; d’altra parte, stimare il numero delle diverse specie di particelle prodotte in una collisione può essere estremamente complicato. In questa tesi, il numero di barioni Lc prodotti in un campione di dati è stato ottenuto utilizzando delle tecniche di machine learning, in grado di apprendere pattern e imparare a distinguere candidate di segnale da quelle di fondo. Si sono inoltre confrontate tre diverse implementazioni di un algoritmo di Boosted Decision Trees (BDT) e si è utilizzata quella più performante per ricostruire il barione Lc in collisioni pp raccolte dall’esperimento ALICE.

Percezione depth guidata mediante proiezione virtuale di pattern

L’utilizzo di informazioni di profondità è oggi di fondamentale utilità per molteplici settori applicativi come la robotica, la guida autonoma o assistita, la realtà aumentata e il monitoraggio ambientale. I sensori di profondità disponibili possono essere divisi in attivi e passivi, dove i sensori passivi ricavano le informazioni di profondità dall'ambiente senza emettere segnali, bensì utilizzando i segnali provenienti dall'ambiente (e.g., luce solare). Nei sensori depth passivi stereo è richiesto un algoritmo per elaborare le immagini delle due camere: la tecnica di stereo matching viene utilizzata appunto per stimare la profondità di una scena. Di recente la ricerca si è occupata anche della sinergia con sensori attivi al fine di migliorare la stima della depth ottenuta da un sensore stereo: si utilizzano i punti affidabili generati dal sensore attivo per guidare l'algoritmo di stereo matching verso la soluzione corretta. In questa tesi si è deciso di affrontare questa tematica da un punto di vista nuovo, utilizzando un sistema di proiezione virtuale di punti corrispondenti in immagini stereo: i pixel delle immagini vengono alterati per guidare l'algoritmo ottimizzando i costi. Un altro vantaggio della strategia proposta è la possibilità di iterare il processo, andando a cambiare il pattern in ogni passo: aggregando i passi in un unico risultato, è possibile migliorare il risultato finale. I punti affidabili sono ottenuti mediante sensori attivi (e.g. LiDAR, ToF), oppure direttamente dalle immagini, stimando la confidenza delle mappe prodotte dal medesimo sistema stereo: la confidenza permette di classificare la bontà di un punto fornito dall'algoritmo di matching. Nel corso della tesi sono stati utilizzati sensori attivi per verificare l'efficacia della proiezione virtuale, ma sono state anche effettuate analisi sulle misure di confidenza: lo scopo è verificare se le misure di confidenza possono rimpiazzare o assistere i sensori attivi.