Deep Learning Architectures for time of arrival detection in Acoustic Emissions Monitoring

Acoustic Emission (AE) monitoring can be used to detect the presence of damage as well as determine its location in Structural Health Monitoring (SHM) applications. Information on the time difference of the signal generated by the damage event arriving at different sensors is essential in performing localization. This makes the time of arrival (ToA) an important piece of information to retrieve from the AE signal. Generally, this is determined using statistical methods such as the Akaike Information Criterion (AIC) which is particularly prone to errors in the presence of noise. And given that the structures of interest are surrounded with harsh environments, a way to accurately estimate the arrival time in such noisy scenarios is of particular interest. In this work, two new methods are presented to estimate the arrival times of AE signals which are based on Machine Learning. Inspired by great results in the field, two models are presented which are Deep Learning models - a subset of machine learning. They are based on Convolutional Neural Network (CNN) and Capsule Neural Network (CapsNet). The primary advantage of such models is that they do not require the user to pre-define selected features but only require raw data to be given and the models establish non-linear relationships between the inputs and outputs. The performance of the models is evaluated using AE signals generated by a custom ray-tracing algorithm by propagating them on an aluminium plate and compared to AIC. It was found that the relative error in estimation on the test set was < 5% for the models compared to around 45% of AIC. The testing process was further continued by preparing an experimental setup and acquiring real AE signals to test on. Similar performances were observed where the two models not only outperform AIC by more than a magnitude in their average errors but also they were shown to be a lot more robust as compared to AIC which fails in the presence of noise.

Machine Learning Unsupervised Methods in the Design of an On-board Health Monitoring System for Satellite Applications

The dissertation starts by providing a description of the phenomena related to the increasing importance recently acquired by satellite applications. The spread of such technology comes with implications, such as an increase in maintenance cost, from which derives the interest in developing advanced techniques that favor an augmented autonomy of spacecrafts in health monitoring. Machine learning techniques are widely employed to lay a foundation for effective systems specialized in fault detection by examining telemetry data. Telemetry consists of a considerable amount of information; therefore, the adopted algorithms must be able to handle multivariate data while facing the limitations imposed by on-board hardware features. In the framework of outlier detection, the dissertation addresses the topic of unsupervised machine learning methods. In the unsupervised scenario, lack of prior knowledge of the data behavior is assumed. In the specific, two models are brought to attention, namely Local Outlier Factor and One-Class Support Vector Machines. Their performances are compared in terms of both the achieved prediction accuracy and the equivalent computational cost. Both models are trained and tested upon the same sets of time series data in a variety of settings, finalized at gaining insights on the effect of the increase in dimensionality. The obtained results allow to claim that both models, combined with a proper tuning of their characteristic parameters, successfully comply with the role of outlier detectors in multivariate time series data. Nevertheless, under this specific context, Local Outlier Factor results to be outperforming One-Class SVM, in that it proves to be more stable over a wider range of input parameter values. This property is especially valuable in unsupervised learning since it suggests that the model is keen to adapting to unforeseen patterns.

Novel Machine-Learning Based IC Testing Strategy

The final goal of the thesis should be a real-world application in the production test data environment. This includes the pre-processing of the data, building models and visualizing the results. To do this, different machine learning models, outlier prediction oriented, should be investigated using a real dataset. Finally, the different outlier prediction algorithms should be compared, and their performance discussed.

End-to-end Deep Metric Learning con Vision-Language Model per il Fashion Image Captioning

L'image captioning è un task di machine learning che consiste nella generazione di una didascalia, o caption, che descriva le caratteristiche di un'immagine data in input. Questo può essere applicato, ad esempio, per descrivere in dettaglio i prodotti in vendita su un sito di e-commerce, migliorando l'accessibilità del sito web e permettendo un acquisto più consapevole ai clienti con difficoltà visive. La generazione di descrizioni accurate per gli articoli di moda online è importante non solo per migliorare le esperienze di acquisto dei clienti, ma anche per aumentare le vendite online. Oltre alla necessità di presentare correttamente gli attributi degli articoli, infatti, descrivere i propri prodotti con il giusto linguaggio può contribuire a catturare l'attenzione dei clienti. In questa tesi, ci poniamo l'obiettivo di sviluppare un sistema in grado di generare una caption che descriva in modo dettagliato l'immagine di un prodotto dell'industria della moda dato in input, sia esso un capo di vestiario o un qualche tipo di accessorio. A questo proposito, negli ultimi anni molti studi hanno proposto soluzioni basate su reti convoluzionali e LSTM. In questo progetto proponiamo invece un'architettura encoder-decoder, che utilizza il modello Vision Transformer per la codifica delle immagini e GPT-2 per la generazione dei testi. Studiamo inoltre come tecniche di deep metric learning applicate in end-to-end durante l'addestramento influenzino le metriche e la qualità delle caption generate dal nostro modello.

Continual learning for computer vision applications

One of the most visionary goals of Artificial Intelligence is to create a system able to mimic and eventually surpass the intelligence observed in biological systems including, ambitiously, the one observed in humans. The main distinctive strength of humans is their ability to build a deep understanding of the world by learning continuously and drawing from their experiences. This ability, which is found in various degrees in all intelligent biological beings, allows them to adapt and properly react to changes by incrementally expanding and refining their knowledge. Arguably, achieving this ability is one of the main goals of Artificial Intelligence and a cornerstone towards the creation of intelligent artificial agents. Modern Deep Learning approaches allowed researchers and industries to achieve great advancements towards the resolution of many long-standing problems in areas like Computer Vision and Natural Language Processing. However, while this current age of renewed interest in AI allowed for the creation of extremely useful applications, a concerningly limited effort is being directed towards the design of systems able to learn continuously. The biggest problem that hinders an AI system from learning incrementally is the catastrophic forgetting phenomenon. This phenomenon, which was discovered in the 90s, naturally occurs in Deep Learning architectures where classic learning paradigms are applied when learning incrementally from a stream of experiences. This dissertation revolves around the Continual Learning field, a sub-field of Machine Learning research that has recently made a comeback following the renewed interest in Deep Learning approaches. This work will focus on a comprehensive view of continual learning by considering algorithmic, benchmarking, and applicative aspects of this field. This dissertation will also touch on community aspects such as the design and creation of research tools aimed at supporting Continual Learning research, and the theoretical and practical aspects concerning public competitions in this field.

Directly training spiking neural networks for cyber-physical systems: from supervised to reinforcement learning

Spiking Neural Networks (SNNs) are bio-inspired Artificial Neural Networks (ANNs) utilizing discrete spiking signals, akin to neuron communication in the brain, making them ideal for real-time and energy-efficient Cyber-Physical Systems (CPSs). This thesis explores their potential in Structural Health Monitoring (SHM), leveraging low-cost MEMS accelerometers for early damage detection in motorway bridges. The study focuses on Long Short-Term SNNs (LSNNs), although their complex learning processes pose challenges. Comparing LSNNs with other ANN models and training algorithms for SHM, findings indicate LSNNs' effectiveness in damage identification, comparable to ANNs trained using traditional methods. Additionally, an optimized embedded LSNN implementation demonstrates a 54% reduction in execution time, but with longer pre-processing due to spike-based encoding. Furthermore, SNNs are applied in UAV obstacle avoidance, trained directly using a Reinforcement Learning (RL) algorithm with event-based input from a Dynamic Vision Sensor (DVS). Performance evaluation against Convolutional Neural Networks (CNNs) highlights SNNs' superior energy efficiency, showing a 6x decrease in energy consumption. The study also investigates embedded SNN implementations' latency and throughput in real-world deployments, emphasizing their potential for energy-efficient monitoring systems. This research contributes to advancing SHM and UAV obstacle avoidance through SNNs' efficient information processing and decision-making capabilities within CPS domains.

Methods for monitoring Machine Learning models

The following thesis aims to investigate the issues concerning the maintenance of a Machine Learning model over time, both about the versioning of the model itself and the data on which it is trained and about data monitoring tools and their distribution. The themes of Data Drift and Concept Drift were then explored and the performance of some of the most popular techniques in the field of Anomaly detection, such as VAE, PCA, and Monte Carlo Dropout, were evaluated.

Flicker perimetry in healthy subjects: influence of age and gender, learning effect and short-term fluctuation

PURPOSE: To determine the mean critical fusion frequency and the short-term fluctuation, to analyze the influence of age, gender, and the learning effect in healthy subjects undergoing flicker perimetry. METHODS: Study 1 - 95 healthy subjects underwent flicker perimetry once in one eye. Mean critical fusion frequency values were compared between genders, and the influence of age was evaluated using linear regression analysis. Study 2 - 20 healthy subjects underwent flicker perimetry 5 times in one eye. The first 3 sessions were separated by an interval of 1 to 30 days, whereas the last 3 sessions were performed within the same day. The first 3 sessions were used to investigate the presence of a learning effect, whereas the last 3 tests were used to calculate short-term fluctuation. RESULTS: Study 1 - Linear regression analysis demonstrated that mean global, foveal, central, and critical fusion frequency per quadrant significantly decreased with age (p<0.05).There were no statistically significant differences in mean critical fusion frequency values between males and females (p>0.05), with the exception of the central area and inferonasal quadrant (p=0.049 and p=0.011, respectively), where the values were lower in females. Study 2 - Mean global (p=0.014), central (p=0.008), and peripheral (p=0.03) critical fusion frequency were significantly lower in the first session compared to the second and third sessions. The mean global short-term fluctuation was 5.06±1.13 Hz, the mean interindividual and intraindividual variabilities were 11.2±2.8% and 6.4±1.5%, respectively. CONCLUSION: This study suggests that, in healthy subjects, critical fusion frequency decreases with age, that flicker perimetry is associated with a learning effect, and that a moderately high short-term fluctuation is expected.

Gastroenterology training in Latin America

Latin America is characterized by ethnic, geographical, cultural, and economic diversity; therefore, training in gastroenterology in the region must be considered in this context. The continent's medical education is characterized by a lack of standards and the volume of research continues to be relatively small. There is a multiplicity of events in general gastroenterology and in sub-disciplines, both at regional and local levels, which ensure that many colleagues have access to information. Medical education programs must be based on a clinical vision and be considered in close contact with the patients. The programs should be properly supervised, appropriately defined, and evaluated on a regular basis. The disparity between the patients' needs, the scarce resources available, and the pressures exerted by the health systems on doctors are frequent cited by those complaining of poor professionalism. Teaching development can play a critical role in ensuring the quality of teaching and learning in universities. Continuing professional development programs activities must be planned on the basis of the doctors' needs, with clearly defined objectives and using proper learning methodologies designed for adults. They must be evaluated and accredited by a competent body, so that they may become the basis of a professional regulatory system. The specialty has made progress in the last decades, offering doctors various possibilities for professional development. The world gastroenterology organization has contributed to the speciality through three distinctive, but closely inter-related, programs: Training Centers, Train-the-Trainers, and Global Guidelines, in which Latin America is deeply involved. (C) 2011 Baishideng. All rights reserved.

Assessment of Preclinical Learning on Oral Surgery Using Three Instructional Strategies

The purpose of this investigation was to evaluate three learning methods for teaching basic oral surgical skills Thirty predoctoral dental students without any surgical knowledge or previous surgical experience were divided Into three groups (n=10 each) according to instructional strategy Group 1, active learning Group 2, text reading only, and Group 3, text reading and video demonstration After instruction, the apprentices were allowed to practice incision dissection and suture maneuvers in a bench learning model During the students' performance, a structured practice evaluation test to account for correct or incorrect maneuvers was applied by trained observers Evaluation tests were repeated after thirty and sixty days Data from resulting scores between groups and periods were considered for statistical analysis (ANOVA and Tukey Kramer) with a significant level of a=0 05 Results showed that the active learning group presented the significantly best learning outcomes related to immediate assimilation of surgical procedures compared to other groups All groups results were similar after sixty days of the first practice Assessment tests were fundamental to evaluate teaching strategies and allowed theoretical and proficiency learning feedbacks Repetition and interactive practice promoted retention of knowledge on basic oral surgical skills

The case of thyroid hormones: how to learn physiology by solving a detective case

Lellis-Santos C, Giannocco G, Nunes MT. The case of thyroid hormones: how to learn physiology by solving a detective case. Adv Physiol Educ 35: 219-226, 2011; doi:10.1152/advan.00135.2010.Thyroid diseases are prevalent among endocrine disorders, and careful evaluation of patients' symptoms is a very important part in their diagnosis. Developing new pedagogical strategies, such as problem-based learning (PBL), is extremely important to stimulate and encourage medical and biomedical students to learn thyroid physiology and identify the signs and symptoms of thyroid dysfunction. The present study aimed to create a new pedagogical approach to build deep knowledge about hypo-/hyperthyroidism by proposing a hands-on activity based on a detective case, using alternative materials in place of laboratory animals. After receiving a description of a criminal story involving changes in thyroid hormone economy, students collected data from clues, such as body weight, mesenteric vascularization, visceral fat, heart and thyroid size, heart rate, and thyroid-stimulating hormone serum concentration to solve the case. Nevertheless, there was one missing clue for each panel of data. Four different materials were proposed to perform the same practical lesson. Animals, pictures, small stuffed toy rats, and illustrations were all effective to promote learning, and the detective case context was considered by students as inviting and stimulating. The activity can be easily performed independently of the institution's purchasing power. The practical lesson stimulated the scientific method of data collection and organization, discussion, and review of thyroid hormone actions to solve the case. Hence, this activity provides a new strategy and alternative materials to teach without animal euthanization.

Intercepting moving targets: does memory from practice in a specific condition of target displacement affect movement timing?

This investigation aimed at assessing the extent to which memory from practice in a specific condition of target displacement modulates temporal errors and movement timing of interceptive movements. We compared two groups practicing with certainty of future target velocity either in unchanged target velocity or in target velocity decrease. Following practice, both experimental groups were probed in the situations of unchanged target velocity and target velocity decrease either under the context of certainty or uncertainty about target velocity. Results from practice showed similar improvement of temporal accuracy between groups, revealing that target velocity decrease did not disturb temporal movement organization when fully predictable. Analysis of temporal errors in the probing trials indicated that both groups had higher timing accuracy in velocity decrease in comparison with unchanged velocity. Effect of practice was detected by increased temporal accuracy of the velocity decrease group in situations of decreased velocity; a trend consistent with the expected effect of practice was observed for temporal errors in the unchanged velocity group and in movement initiation at a descriptive level. An additional point of theoretical interest was the fast adaptation in both groups to a target velocity pattern different from that practiced. These points are discussed under the perspective of integration of vision and motor control by means of an internal forward model of external motion.

Harmonic Distortion State Estimation Using an Evolutionary Strategy

This paper presents a new methodology to estimate harmonic distortions in a power system, based on measurements of a limited number of given sites. The algorithm utilizes evolutionary strategies (ES), a development branch of evolutionary algorithms. The main advantage in using such a technique relies upon its modeling facilities as well as its potential to solve fairly complex problems. The problem-solving algorithm herein proposed makes use of data from various power-quality (PQ) meters, which can either be synchronized by high technology global positioning system devices or by using information from a fundamental frequency load flow. This second approach makes the overall PQ monitoring system much less costly. The algorithm is applied to an IEEE test network, for which sensitivity analysis is performed to determine how the parameters of the ES can be selected so that the algorithm performs in an effective way. Case studies show fairly promising results and the robustness of the proposed method.

Fundamental Skills in Physics: Embedded Learning

Student attitudes towards a subject affect their learning. For students in physics service courses, relevance is emphasised by vocational applications. A similar strategy is being used for students who aspire to continued study of physics, in an introduction to fundamental skills in experimental physics – the concepts, computational tools and practical skills involved in appropriately obtaining and interpreting measurement data. An educational module is being developed that aims to enhance the student experience by embedding learning of these skills in the practicing physicist’s activity of doing an experiment (gravity estimation using a rolling pendulum). The group concentrates on particular skills prompted by challenges such as: • How can we get an answer to our question? • How good is our answer? • How can it be improved? This explicitly provides students the opportunity to consider and construct their own ideas. It gives them time to discuss, digest and practise without undue stress, thereby assisting them to internalise core skills. Design of the learning activity is approached in an iterative manner, via theoretical and practical considerations, with input from a range of teaching staff, and subject to trials of prototypes.

The interorganizational virtual organization: Defining a Weberian ideal

The concept of the virtual organization (VO) has engendered great interest in the literature, yet there is still little common understanding of the concept, as evidenced by the multitude of labels applied to VOs. In this article, we focus on a “Weberian-ideal-type” definition of the interorganizational VO, posited in our earlier work (Kasper-Fuehrer and Ashkanasy 2001). We argue, however, that this definition left unanswered critical questions relating to the nature and effects of interorganizational VOs. We answer these questions here by explicating the terms in the definition and deriving ten corollaries, or “natural consequences” of our definition. The corollaries posit that interorganizational VOs are temporary in nature, are network organizations, are independent, and are based on swift trust. We suggest further that interorganizational VOs enable small to medium enterprises to exploit market opportunities, and enable VO member organizations to create a value-adding partnership. We also identify information and communication technology (ICT) as the essential enabler of VOs. Finally, we argue that interorganizational VOs act as a single organizational unit and that they therefore constitute a uniquely distinguishable organizational form. We conclude with suggestions for further research, including trust, organizational behavior, transaction economics, virtual HRM, and business strategy.