Simulating event related potentials in tasks involving conflict resolution: a neural network

In this thesis, the main Executive Control theories are exposed. Methods typical of Cognitive and Computational Neuroscience are introduced and the role of behavioural tasks involving conflict resolution in the response elaboration, after the presentation of a stimulus to the subject, are highlighted. In particular, the Eriksen Flanker Task and its variants are discussed. Behavioural data, from scientific literature, are illustrated in terms of response times and error rates. During experimental behavioural tasks, EEG is registered simultaneously. Thanks to this, event related potential, related with the current task, can be studied. Different theories regarding relevant event related potential in this field - such as N2, fERN (feedback Error Related Negativity) and ERN (Error Related Negativity) – are introduced. The aim of this thesis is to understand and simulate processes regarding Executive Control, including performance improvement, error detection mechanisms, post error adjustments and the role of selective attention, with the help of an original neural network model. The network described here has been built with the purpose to simulate behavioural results of a four choice Eriksen Flanker Task. Model results show that the neural network can simulate response times, error rates and event related potentials quite well. Finally, results are compared with behavioural data and discussed in light of the mentioned Executive Control theories. Future perspective for this new model are outlined.

Self-supervised information retrieval: a novel approach based on Deep Metric Learning and Neural Language Models

Most of the existing open-source search engines, utilize keyword or tf-idf based techniques to find relevant documents and web pages relative to an input query. Although these methods, with the help of a page rank or knowledge graphs, proved to be effective in some cases, they often fail to retrieve relevant instances for more complicated queries that would require a semantic understanding to be exploited. In this Thesis, a self-supervised information retrieval system based on transformers is employed to build a semantic search engine over the library of Gruppo Maggioli company. Semantic search or search with meaning can refer to an understanding of the query, instead of simply finding words matches and, in general, it represents knowledge in a way suitable for retrieval. We chose to investigate a new self-supervised strategy to handle the training of unlabeled data based on the creation of pairs of ’artificial’ queries and the respective positive passages. We claim that by removing the reliance on labeled data, we may use the large volume of unlabeled material on the web without being limited to languages or domains where labeled data is abundant.

Investigating perceptual multisensory impairments in Autism through a neural network: a possible neural implementation for the shift from cross-modal competition to facilitation

Resolution of multisensory deficits has been observed in teenagers with Autism Spectrum Disorders (ASD) for complex, social speech stimuli; this resolution extends to more basic multisensory processing, involving low-level stimuli. In particular, a delayed transition of multisensory integration (MSI) from a default state of competition to one of facilitation has been observed in ASD children. In other terms, the complete maturation of MSI is achieved later in ASD. In the present study a neuro-computational model is used to reproduce some patterns of behavior observed experimentally, modeling a bisensory reaction time task, in which auditory and visual stimuli are presented in random sequence alone (A or V) or together (AV). The model explains how the default competitive state can be implemented via mutual inhibition between primary sensory areas, and how the shift toward the classical multisensory facilitation, observed in adults, is the result of inhibitory cross-modal connections becoming excitatory during the development. Model results are consistent with a stronger cross-modal inhibition in ASD children, compared to normotypical (NT) ones, suggesting that the transition toward a cooperative interaction between sensory modalities takes longer to occur. Interestingly, the model also predicts the difference between unisensory switch trials (in which sensory modality switches) and unisensory repeat trials (in which sensory modality repeats). This is due to an inhibitory mechanism, characterized by a slow dynamics, driven by the preceding stimulus and inhibiting the processing of the incoming one, when of the opposite sensory modality. These findings link the cognitive framework delineated by the empirical results to a plausible neural implementation.

Neural Network based Non Orthogonal Random Access for 6G NTN-IoT

Pervasive and distributed Internet of Things (IoT) devices demand ubiquitous coverage beyond No-man’s land. To satisfy plethora of IoT devices with resilient connectivity, Non-Terrestrial Networks (NTN) will be pivotal to assist and complement terrestrial systems. In a massiveMTC scenario over NTN, characterized by sporadic uplink data reports, all the terminals within a satellite beam shall be served during the short visibility window of the flying platform, thus generating congestion due to simultaneous access attempts of IoT devices on the same radio resource. The more terminals collide, the more average-time it takes to complete an access which is due to the decreased number of successful attempts caused by Back-off commands of legacy methods. A possible countermeasure is represented by Non-Orthogonal Multiple Access scheme, which requires the knowledge of the number of superimposed NPRACH preambles. This work addresses this problem by proposing a Neural Network (NN) algorithm to cope with the uncoordinated random access performed by a prodigious number of Narrowband-IoT devices. Our proposed method classifies the number of colliding users, and for each estimates the Time of Arrival (ToA). The performance assessment, under Line of Sight (LoS) and Non-LoS conditions in sub-urban environments with two different satellite configurations, shows significant benefits of the proposed NN algorithm with respect to traditional methods for the ToA estimation.

Tactile perception - Perception of tactile distance changes with body site: a neural network modelling study.

La distorsione della percezione della distanza tra due stimoli puntuali applicati sulla superfice della pelle di diverse regioni corporee è conosciuta come Illusione di Weber. Questa illusione è stata osservata, e verificata, in molti esperimenti in cui ai soggetti era chiesto di giudicare la distanza tra due stimoli applicati sulla superficie della pelle di differenti parti corporee. Da tali esperimenti si è dedotto che una stessa distanza tra gli stimoli è giudicata differentemente per diverse regioni corporee. Il concetto secondo cui la distanza sulla pelle è spesso percepita in maniera alterata è ampiamente condiviso, ma i meccanismi neurali che manovrano questa illusione sono, allo stesso tempo, ancora ampiamente sconosciuti. In particolare, non è ancora chiaro come sia interpretata la distanza tra due stimoli puntuali simultanei, e quali aree celebrali siano coinvolte in questa elaborazione. L’illusione di Weber può essere spiegata, in parte, considerando la differenza in termini di densità meccano-recettoriale delle differenti regioni corporee, e l’immagine distorta del nostro corpo che risiede nella Corteccia Primaria Somato-Sensoriale (homunculus). Tuttavia, questi meccanismi sembrano non sufficienti a spiegare il fenomeno osservato: infatti, secondo i risultati derivanti da 100 anni di sperimentazioni, le distorsioni effettive nel giudizio delle distanze sono molto più piccole rispetto alle distorsioni che la Corteccia Primaria suggerisce. In altre parole, l’illusione osservata negli esperimenti tattili è molto più piccola rispetto all’effetto prodotto dalla differente densità recettoriale che affligge le diverse parti del corpo, o dall’estensione corticale. Ciò, ha portato a ipotizzare che la percezione della distanza tattile richieda la presenza di un’ulteriore area celebrale, e di ulteriori meccanismi che operino allo scopo di ridimensionare – almeno parzialmente – le informazioni derivanti dalla corteccia primaria, in modo da mantenere una certa costanza nella percezione della distanza tattile lungo la superfice corporea. E’ stata così proposta la presenza di una sorta di “processo di ridimensionamento”, chiamato “Rescaling Process” che opera per ridurre questa illusione verso una percezione più verosimile. Il verificarsi di questo processo è sostenuto da molti ricercatori in ambito neuro scientifico; in particolare, dal Dr. Matthew Longo, neuro scienziato del Department of Psychological Sciences (Birkbeck University of London), le cui ricerche sulla percezione della distanza tattile e sulla rappresentazione corporea sembrano confermare questa ipotesi. Tuttavia, i meccanismi neurali, e i circuiti che stanno alla base di questo potenziale “Rescaling Process” sono ancora ampiamente sconosciuti. Lo scopo di questa tesi è stato quello di chiarire la possibile organizzazione della rete, e i meccanismi neurali che scatenano l’illusione di Weber e il “Rescaling Process”, usando un modello di rete neurale. La maggior parte del lavoro è stata svolta nel Dipartimento di Scienze Psicologiche della Birkbeck University of London, sotto la supervisione del Dott. M. Longo, il quale ha contribuito principalmente all’interpretazione dei risultati del modello, dando suggerimenti sull’elaborazione dei risultati in modo da ottenere un’informazione più chiara; inoltre egli ha fornito utili direttive per la validazione dei risultati durante l’implementazione di test statistici. Per replicare l’illusione di Weber ed il “Rescaling Proess”, la rete neurale è stata organizzata con due strati principali di neuroni corrispondenti a due differenti aree funzionali corticali: • Primo strato di neuroni (il quale dà il via ad una prima elaborazione degli stimoli esterni): questo strato può essere pensato come parte della Corteccia Primaria Somato-Sensoriale affetta da Magnificazione Corticale (homunculus). • Secondo strato di neuroni (successiva elaborazione delle informazioni provenienti dal primo strato): questo strato può rappresentare un’Area Corticale più elevata coinvolta nell’implementazione del “Rescaling Process”. Le reti neurali sono state costruite includendo connessioni sinaptiche all’interno di ogni strato (Sinapsi Laterali), e connessioni sinaptiche tra i due strati neurali (Sinapsi Feed-Forward), assumendo inoltre che l’attività di ogni neurone dipenda dal suo input attraverso una relazione sigmoidale statica, cosi come da una dinamica del primo ordine. In particolare, usando la struttura appena descritta, sono state implementate due differenti reti neurali, per due differenti regioni corporee (per esempio, Mano e Braccio), caratterizzate da differente risoluzione tattile e differente Magnificazione Corticale, in modo da replicare l’Illusione di Weber ed il “Rescaling Process”. Questi modelli possono aiutare a comprendere il meccanismo dell’illusione di Weber e dare così una possibile spiegazione al “Rescaling Process”. Inoltre, le reti neurali implementate forniscono un valido contributo per la comprensione della strategia adottata dal cervello nell’interpretazione della distanza sulla superficie della pelle. Oltre allo scopo di comprensione, tali modelli potrebbero essere impiegati altresì per formulare predizioni che potranno poi essere verificate in seguito, in vivo, su soggetti reali attraverso esperimenti di percezione tattile. E’ importante sottolineare che i modelli implementati sono da considerarsi prettamente come modelli funzionali e non intendono replicare dettagli fisiologici ed anatomici. I principali risultati ottenuti tramite questi modelli sono la riproduzione del fenomeno della “Weber’s Illusion” per due differenti regioni corporee, Mano e Braccio, come riportato nei tanti articoli riguardanti le illusioni tattili (per esempio “The perception of distance and location for dual tactile pressures” di Barry G. Green). L’illusione di Weber è stata registrata attraverso l’output delle reti neurali, e poi rappresentata graficamente, cercando di spiegare le ragioni di tali risultati.

Neural Network Based UAV Fault Tolerant Control with Aerodynamic Disturbance Estimation and Compensation

In this thesis, the problem of controlling a quadrotor UAV is considered. It is done by presenting an original control system, designed as a combination of Neural Networks and Disturbance Observer, using a composite learning approach for a system of the second order, which is a novel methodology in literature. After a brief introduction about the quadrotors, the concepts needed to understand the controller are presented, such as the main notions of advanced control, the basic structure and design of a Neural Network, the modeling of a quadrotor and its dynamics. The full simulator, developed on the MATLAB Simulink environment, used throughout the whole thesis, is also shown. For the guidance and control purposes, a Sliding Mode Controller, used as a reference, it is firstly introduced, and its theory and implementation on the simulator are illustrated. Finally the original controller is introduced, through its novel formulation, and implementation on the model. The effectiveness and robustness of the two controllers are then proven by extensive simulations in all different conditions of external disturbance and faults.

Convolutional Neural Network Architectures for Template Matching

The Neural Networks customized and tested in this thesis (WaldoNet, FlowNet and PatchNet) are a first exploration and approach to the Template Matching task. The possibilities of extension are therefore many and some are proposed below. During my thesis, I have analyzed the functioning of the classical algorithms and adapted with deep learning algorithms. The features extracted from both the template and the query images resemble the keypoints of the SIFT algorithm. Then, instead of similarity function or keypoints matching, WaldoNet and PatchNet use the convolutional layer to compare the features, while FlowNet uses the correlational layer. In addition, I have identified the major challenges of the Template Matching task (affine/non-affine transformations, intensity changes...) and solved them with a careful design of the dataset.

Production Lot Detection through Synthetic Generated Data

Correctness of information gathered in production environments is an essential part of quality assurance processes in many industries, this task is often performed by human resources who visually take annotations in various steps of the production flow. Depending on the performed task the correlation between where exactly the information is gathered and what it represents is more than often lost in the process. The lack of labeled data places a great boundary on the application of deep neural networks aimed at object detection tasks, moreover supervised training of deep models requires a great amount of data to be available. Reaching an adequate large collection of labeled images through classic techniques of data annotations is an exhausting and costly task to perform, not always suitable for every scenario. A possible solution is to generate synthetic data that replicates the real one and use it to fine-tune a deep neural network trained on one or more source domains to a different target domain. The purpose of this thesis is to show a real case scenario where the provided data were both in great scarcity and missing the required annotations. Sequentially a possible approach is presented where synthetic data has been generated to address those issues while standing as a training base of deep neural networks for object detection, capable of working on images taken in production-like environments. Lastly, it compares performance on different types of synthetic data and convolutional neural networks used as backbones for the model.

Deep learning for computer vision: a comparison between convolutional neural networks and hierarchical temporal memories on object recognition tasks

In recent years, Deep Learning techniques have shown to perform well on a large variety of problems both in Computer Vision and Natural Language Processing, reaching and often surpassing the state of the art on many tasks. The rise of deep learning is also revolutionizing the entire field of Machine Learning and Pattern Recognition pushing forward the concepts of automatic feature extraction and unsupervised learning in general. However, despite the strong success both in science and business, deep learning has its own limitations. It is often questioned if such techniques are only some kind of brute-force statistical approaches and if they can only work in the context of High Performance Computing with tons of data. Another important question is whether they are really biologically inspired, as claimed in certain cases, and if they can scale well in terms of "intelligence". The dissertation is focused on trying to answer these key questions in the context of Computer Vision and, in particular, Object Recognition, a task that has been heavily revolutionized by recent advances in the field. Practically speaking, these answers are based on an exhaustive comparison between two, very different, deep learning techniques on the aforementioned task: Convolutional Neural Network (CNN) and Hierarchical Temporal memory (HTM). They stand for two different approaches and points of view within the big hat of deep learning and are the best choices to understand and point out strengths and weaknesses of each of them. CNN is considered one of the most classic and powerful supervised methods used today in machine learning and pattern recognition, especially in object recognition. CNNs are well received and accepted by the scientific community and are already deployed in large corporation like Google and Facebook for solving face recognition and image auto-tagging problems. HTM, on the other hand, is known as a new emerging paradigm and a new meanly-unsupervised method, that is more biologically inspired. It tries to gain more insights from the computational neuroscience community in order to incorporate concepts like time, context and attention during the learning process which are typical of the human brain. In the end, the thesis is supposed to prove that in certain cases, with a lower quantity of data, HTM can outperform CNN.

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.

Edge-aware Graph Attention Networks: Joint Reasoning On Text and Knowledge Graphs for Biomedical Question Answering

Much of the real-world dataset, including textual data, can be represented using graph structures. The use of graphs to represent textual data has many advantages, mainly related to maintaining a more significant amount of information, such as the relationships between words and their types. In recent years, many neural network architectures have been proposed to deal with tasks on graphs. Many of them consider only node features, ignoring or not giving the proper relevance to relationships between them. However, in many node classification tasks, they play a fundamental role. This thesis aims to analyze the main GNNs, evaluate their advantages and disadvantages, propose an innovative solution considered as an extension of GAT, and apply them to a case study in the biomedical field. We propose the reference GNNs, implemented with methodologies later analyzed, and then applied to a question answering system in the biomedical field as a replacement for the pre-existing GNN. We attempt to obtain better results by using models that can accept as input both node and edge features. As shown later, our proposed models can beat the original solution and define the state-of-the-art for the task under analysis.

Machine learning "as a service" for high energy physics (MLaaS4HEP): evolution of a framework for ML-based physics analyses

The scientific success of the LHC experiments at CERN highly depends on the availability of computing resources which efficiently store, process, and analyse the amount of data collected every year. This is ensured by the Worldwide LHC Computing Grid infrastructure that connect computing centres distributed all over the world with high performance network. LHC has an ambitious experimental program for the coming years, which includes large investments and improvements both for the hardware of the detectors and for the software and computing systems, in order to deal with the huge increase in the event rate expected from the High Luminosity LHC (HL-LHC) phase and consequently with the huge amount of data that will be produced. Since few years the role of Artificial Intelligence has become relevant in the High Energy Physics (HEP) world. Machine Learning (ML) and Deep Learning algorithms have been successfully used in many areas of HEP, like online and offline reconstruction programs, detector simulation, object reconstruction, identification, Monte Carlo generation, and surely they will be crucial in the HL-LHC phase. This thesis aims at contributing to a CMS R&D project, regarding a ML "as a Service" solution for HEP needs (MLaaS4HEP). It consists in a data-service able to perform an entire ML pipeline (in terms of reading data, processing data, training ML models, serving predictions) in a completely model-agnostic fashion, directly using ROOT files of arbitrary size from local or distributed data sources. This framework has been updated adding new features in the data preprocessing phase, allowing more flexibility to the user. Since the MLaaS4HEP framework is experiment agnostic, the ATLAS Higgs Boson ML challenge has been chosen as physics use case, with the aim to test MLaaS4HEP and the contribution done with this work.

Predicting corticospinal excitability from pre-TMS EEG using convolutional neural networks

The amplitude of motor evoked potentials (MEPs) elicited by transcranial magnetic stimulation (TMS) of the primary motor cortex (M1) shows a large variability from trial to trial, although MEPs are evoked by the same repeated stimulus. A multitude of factors is believed to influence MEP amplitudes, such as cortical, spinal and motor excitability state. The goal of this work is to explore to which degree the variation in MEP amplitudes can be explained by the cortical state right before the stimulation. Specifically, we analyzed a dataset acquired on eleven healthy subjects comprising, for each subject, 840 single TMS pulses applied to the left M1 during acquisition of electroencephalography (EEG) and electromyography (EMG). An interpretable convolutional neural network, named SincEEGNet, was utilized to discriminate between low- and high-corticospinal excitability trials, defined according to the MEP amplitude, using in input the pre-TMS EEG. This data-driven approach enabled considering multiple brain locations and frequency bands without any a priori selection. Post-hoc interpretation techniques were adopted to enhance interpretation by identifying the more relevant EEG features for the classification. Results show that individualized classifiers successfully discriminated between low and high M1 excitability states in all participants. Outcomes of the interpretation methods suggest the importance of the electrodes situated over the TMS stimulation site, as well as the relevance of the temporal samples of the input EEG closer to the stimulation time. This novel decoding method allows causal investigation of the cortical excitability state, which may be relevant for personalizing and increasing the efficacy of therapeutic brain-state dependent brain stimulation (for example in patients affected by Parkinson’s disease).

Mitigating non-Lambertian surfaces issues in Stereo Matching with Neural Radiance Fields

Depth estimation from images has long been regarded as a preferable alternative compared to expensive and intrusive active sensors, such as LiDAR and ToF. The topic has attracted the attention of an increasingly wide audience thanks to the great amount of application domains, such as autonomous driving, robotic navigation and 3D reconstruction. Among the various techniques employed for depth estimation, stereo matching is one of the most widespread, owing to its robustness, speed and simplicity in setup. Recent developments has been aided by the abundance of annotated stereo images, which granted to deep learning the opportunity to thrive in a research area where deep networks can reach state-of-the-art sub-pixel precision in most cases. Despite the recent findings, stereo matching still begets many open challenges, two among them being finding pixel correspondences in presence of objects that exhibits a non-Lambertian behaviour and processing high-resolution images. Recently, a novel dataset named Booster, which contains high-resolution stereo pairs featuring a large collection of labeled non-Lambertian objects, has been released. The work shown that training state-of-the-art deep neural network on such data improves the generalization capabilities of these networks also in presence of non-Lambertian surfaces. Regardless being a further step to tackle the aforementioned challenge, Booster includes a rather small number of annotated images, and thus cannot satisfy the intensive training requirements of deep learning. This thesis work aims to investigate novel view synthesis techniques to augment the Booster dataset, with ultimate goal of improving stereo matching reliability in presence of high-resolution images that displays non-Lambertian surfaces.

Tree regularized neural networks for key point analysis

This thesis contributes to the ArgMining 2021 shared task on Key Point Analysis. Key Point Analysis entails extracting and calculating the prevalence of a concise list of the most prominent talking points, from an input corpus. These talking points are usually referred to as key points. Key point analysis is divided into two subtasks: Key Point Matching, which involves assigning a matching score to each key point/argument pair, and Key Point Generation, which consists of the generation of key points. The task of Key Point Matching was approached using different models: a pretrained Sentence Transformers model and a tree-constrained Graph Neural Network were tested. The best model was the fine-tuned Sentence Transformers, which achieved a mean Average Precision score of 0.75, ranking 12 compared to other participating teams. The model was then used for the subtask of Key Point Generation using the extractive method in the selection of key point candidates and the model developed for the previous subtask to evaluate them.