Application of Deep Learning techniques in the search for BSM Higgs bosons in the $\mu\mu$ final state in CMS

The Standard Model (SM) of particle physics predicts the existence of a Higgs field responsible for the generation of particles' mass. However, some aspects of this theory remain unsolved, supposing the presence of new physics Beyond the Standard Model (BSM) with the production of new particles at a higher energy scale compared to the current experimental limits. The search for additional Higgs bosons is, in fact, predicted by theoretical extensions of the SM including the Minimal Supersymmetry Standard Model (MSSM). In the MSSM, the Higgs sector consists of two Higgs doublets, resulting in five physical Higgs particles: two charged bosons $H^{\pm}$, two neutral scalars $h$ and $H$, and one pseudoscalar $A$. The work presented in this thesis is dedicated to the search of neutral non-Standard Model Higgs bosons decaying to two muons in the model independent MSSM scenario. Proton-proton collision data recorded by the CMS experiment at the CERN LHC at a center-of-mass energy of 13 TeV are used, corresponding to an integrated luminosity of $35.9\ \text{fb}^{-1}$. Such search is sensitive to neutral Higgs bosons produced either via gluon fusion process or in association with a $\text{b}\bar{\text{b}}$ quark pair. The extensive usage of Machine and Deep Learning techniques is a fundamental element in the discrimination between signal and background simulated events. A new network structure called parameterised Neural Network (pNN) has been implemented, replacing a whole set of single neural networks trained at a specific mass hypothesis value with a single neural network able to generalise well and interpolate in the entire mass range considered. The results of the pNN signal/background discrimination are used to set a model independent 95\% confidence level expected upper limit on the production cross section times branching ratio, for a generic $\phi$ boson decaying into a muon pair in the 130 to 1000 GeV range.

Development and characterization of deep learning techniques for neuroimaging data

Deep learning methods are extremely promising machine learning tools to analyze neuroimaging data. However, their potential use in clinical settings is limited because of the existing challenges of applying these methods to neuroimaging data. In this study, first a data leakage type caused by slice-level data split that is introduced during training and validation of a 2D CNN is surveyed and a quantitative assessment of the model’s performance overestimation is presented. Second, an interpretable, leakage-fee deep learning software written in a python language with a wide range of options has been developed to conduct both classification and regression analysis. The software was applied to the study of mild cognitive impairment (MCI) in patients with small vessel disease (SVD) using multi-parametric MRI data where the cognitive performance of 58 patients measured by five neuropsychological tests is predicted using a multi-input CNN model taking brain image and demographic data. Each of the cognitive test scores was predicted using different MRI-derived features. As MCI due to SVD has been hypothesized to be the effect of white matter damage, DTI-derived features MD and FA produced the best prediction outcome of the TMT-A score which is consistent with the existing literature. In a second study, an interpretable deep learning system aimed at 1) classifying Alzheimer disease and healthy subjects 2) examining the neural correlates of the disease that causes a cognitive decline in AD patients using CNN visualization tools and 3) highlighting the potential of interpretability techniques to capture a biased deep learning model is developed. Structural magnetic resonance imaging (MRI) data of 200 subjects was used by the proposed CNN model which was trained using a transfer learning-based approach producing a balanced accuracy of 71.6%. Brain regions in the frontal and parietal lobe showing the cerebral cortex atrophy were highlighted by the visualization tools.

Machine learning methods for prediction of disulphide bonding states of cysteine residues in proteins

The goal of this thesis work is to develop a computational method based on machine learning techniques for predicting disulfide-bonding states of cysteine residues in proteins, which is a sub-problem of a bigger and yet unsolved problem of protein structure prediction. Improvement in the prediction of disulfide bonding states of cysteine residues will help in putting a constraint in the three dimensional (3D) space of the respective protein structure, and thus will eventually help in the prediction of 3D structure of proteins. Results of this work will have direct implications in site-directed mutational studies of proteins, proteins engineering and the problem of protein folding. We have used a combination of Artificial Neural Network (ANN) and Hidden Markov Model (HMM), the so-called Hidden Neural Network (HNN) as a machine learning technique to develop our prediction method. By using different global and local features of proteins (specifically profiles, parity of cysteine residues, average cysteine conservation, correlated mutation, sub-cellular localization, and signal peptide) as inputs and considering Eukaryotes and Prokaryotes separately we have reached to a remarkable accuracy of 94% on cysteine basis for both Eukaryotic and Prokaryotic datasets, and an accuracy of 90% and 93% on protein basis for Eukaryotic dataset and Prokaryotic dataset respectively. These accuracies are best so far ever reached by any existing prediction methods, and thus our prediction method has outperformed all the previously developed approaches and therefore is more reliable. Most interesting part of this thesis work is the differences in the prediction performances of Eukaryotes and Prokaryotes at the basic level of input coding when ‘profile’ information was given as input to our prediction method. And one of the reasons for this we discover is the difference in the amino acid composition of the local environment of bonded and free cysteine residues in Eukaryotes and Prokaryotes. Eukaryotic bonded cysteine examples have a ‘symmetric-cysteine-rich’ environment, where as Prokaryotic bonded examples lack it.

Learning with Kernels on Graphs: DAG-based kernels, data streams and RNA function prediction.

In many application domains data can be naturally represented as graphs. When the application of analytical solutions for a given problem is unfeasible, machine learning techniques could be a viable way to solve the problem. Classical machine learning techniques are defined for data represented in a vectorial form. Recently some of them have been extended to deal directly with structured data. Among those techniques, kernel methods have shown promising results both from the computational complexity and the predictive performance point of view. Kernel methods allow to avoid an explicit mapping in a vectorial form relying on kernel functions, which informally are functions calculating a similarity measure between two entities. However, the definition of good kernels for graphs is a challenging problem because of the difficulty to find a good tradeoff between computational complexity and expressiveness. Another problem we face is learning on data streams, where a potentially unbounded sequence of data is generated by some sources. There are three main contributions in this thesis. The first contribution is the definition of a new family of kernels for graphs based on Directed Acyclic Graphs (DAGs). We analyzed two kernels from this family, achieving state-of-the-art results from both the computational and the classification point of view on real-world datasets. The second contribution consists in making the application of learning algorithms for streams of graphs feasible. Moreover,we defined a principled way for the memory management. The third contribution is the application of machine learning techniques for structured data to non-coding RNA function prediction. In this setting, the secondary structure is thought to carry relevant information. However, existing methods considering the secondary structure have prohibitively high computational complexity. We propose to apply kernel methods on this domain, obtaining state-of-the-art results.

Machine learning per l'idrologia: applicazione del metodo random forests per la previsione degli eventi di piena fluviale con un approccio probabilistico

Le tecniche di Machine Learning sono molto utili in quanto consento di massimizzare l’utilizzo delle informazioni in tempo reale. Il metodo Random Forests può essere annoverato tra le tecniche di Machine Learning più recenti e performanti. Sfruttando le caratteristiche e le potenzialità di questo metodo, la presente tesi di dottorato affronta due casi di studio differenti; grazie ai quali è stato possibile elaborare due differenti modelli previsionali. Il primo caso di studio si è incentrato sui principali fiumi della regione Emilia-Romagna, caratterizzati da tempi di risposta molto brevi. La scelta di questi fiumi non è stata casuale: negli ultimi anni, infatti, in detti bacini si sono verificati diversi eventi di piena, in gran parte di tipo “flash flood”. Il secondo caso di studio riguarda le sezioni principali del fiume Po, dove il tempo di propagazione dell’onda di piena è maggiore rispetto ai corsi d’acqua del primo caso di studio analizzato. Partendo da una grande quantità di dati, il primo passo è stato selezionare e definire i dati in ingresso in funzione degli obiettivi da raggiungere, per entrambi i casi studio. Per l’elaborazione del modello relativo ai fiumi dell’Emilia-Romagna, sono stati presi in considerazione esclusivamente i dati osservati; a differenza del bacino del fiume Po in cui ai dati osservati sono stati affiancati anche i dati di previsione provenienti dalla catena modellistica Mike11 NAM/HD. Sfruttando una delle principali caratteristiche del metodo Random Forests, è stata stimata una probabilità di accadimento: questo aspetto è fondamentale sia nella fase tecnica che in fase decisionale per qualsiasi attività di intervento di protezione civile. L'elaborazione dei dati e i dati sviluppati sono stati effettuati in ambiente R. Al termine della fase di validazione, gli incoraggianti risultati ottenuti hanno permesso di inserire il modello sviluppato nel primo caso studio all’interno dell’architettura operativa di FEWS.

Challenges and Opportunities of Machine Learning for Clinical and Omics Data

Clinical and omics data are a promising field of application for machine learning techniques even though these methods are not yet systematically adopted in healthcare institutions. Despite artificial intelligence has proved successful in terms of prediction of pathologies or identification of their causes, the systematic adoption of these techniques still presents challenging issues due to the peculiarities of the analysed data. The aim of this thesis is to apply machine learning algorithms to both clinical and omics data sets in order to predict a patient's state of health and get better insights on the possible causes of the analysed diseases. In doing so, many of the arising issues when working with medical data will be discussed while possible solutions will be proposed to make machine learning provide feasible results and possibly become an effective and reliable support tool for healthcare systems.

Methods for integrating machine learning and constrained optimization

In the framework of industrial problems, the application of Constrained Optimization is known to have overall very good modeling capability and performance and stands as one of the most powerful, explored, and exploited tool to address prescriptive tasks. The number of applications is huge, ranging from logistics to transportation, packing, production, telecommunication, scheduling, and much more. The main reason behind this success is to be found in the remarkable effort put in the last decades by the OR community to develop realistic models and devise exact or approximate methods to solve the largest variety of constrained or combinatorial optimization problems, together with the spread of computational power and easily accessible OR software and resources. On the other hand, the technological advancements lead to a data wealth never seen before and increasingly push towards methods able to extract useful knowledge from them; among the data-driven methods, Machine Learning techniques appear to be one of the most promising, thanks to its successes in domains like Image Recognition, Natural Language Processes and playing games, but also the amount of research involved. The purpose of the present research is to study how Machine Learning and Constrained Optimization can be used together to achieve systems able to leverage the strengths of both methods: this would open the way to exploiting decades of research on resolution techniques for COPs and constructing models able to adapt and learn from available data. In the first part of this work, we survey the existing techniques and classify them according to the type, method, or scope of the integration; subsequently, we introduce a novel and general algorithm devised to inject knowledge into learning models through constraints, Moving Target. In the last part of the thesis, two applications stemming from real-world projects and done in collaboration with Optit will be presented.

The three-dimensional single-bin-size bin packing problem: combining metaheuristic and machine learning approaches

The Three-Dimensional Single-Bin-Size Bin Packing Problem is one of the most studied problem in the Cutting & Packing category. From a strictly mathematical point of view, it consists of packing a finite set of strongly heterogeneous “small” boxes, called items, into a finite set of identical “large” rectangles, called bins, minimizing the unused volume and requiring that the items are packed without overlapping. The great interest is mainly due to the number of real-world applications in which it arises, such as pallet and container loading, cutting objects out of a piece of material and packaging design. Depending on these real-world applications, more objective functions and more practical constraints could be needed. After a brief discussion about the real-world applications of the problem and a exhaustive literature review, the design of a two-stage algorithm to solve the aforementioned problem is presented. The algorithm must be able to provide the spatial coordinates of the placed boxes vertices and also the optimal boxes input sequence, while guaranteeing geometric, stability, fragility constraints and a reduced computational time. Due to NP-hard complexity of this type of combinatorial problems, a fusion of metaheuristic and machine learning techniques is adopted. In particular, a hybrid genetic algorithm coupled with a feedforward neural network is used. In the first stage, a rich dataset is created starting from a set of real input instances provided by an industrial company and the feedforward neural network is trained on it. After its training, given a new input instance, the hybrid genetic algorithm is able to run using the neural network output as input parameter vector, providing as output the optimal solution. The effectiveness of the proposed works is confirmed via several experimental tests.

Statistical methods for biomedical signal analysis and processing

Statistical modelling and statistical learning theory are two powerful analytical frameworks for analyzing signals and developing efficient processing and classification algorithms. In this thesis, these frameworks are applied for modelling and processing biomedical signals in two different contexts: ultrasound medical imaging systems and primate neural activity analysis and modelling. In the context of ultrasound medical imaging, two main applications are explored: deconvolution of signals measured from a ultrasonic transducer and automatic image segmentation and classification of prostate ultrasound scans. In the former application a stochastic model of the radio frequency signal measured from a ultrasonic transducer is derived. This model is then employed for developing in a statistical framework a regularized deconvolution procedure, for enhancing signal resolution. In the latter application, different statistical models are used to characterize images of prostate tissues, extracting different features. These features are then uses to segment the images in region of interests by means of an automatic procedure based on a statistical model of the extracted features. Finally, machine learning techniques are used for automatic classification of the different region of interests. In the context of neural activity signals, an example of bio-inspired dynamical network was developed to help in studies of motor-related processes in the brain of primate monkeys. The presented model aims to mimic the abstract functionality of a cell population in 7a parietal region of primate monkeys, during the execution of learned behavioural tasks.

Kernel Methods for Tree Structured Data

Machine learning comprises a series of techniques for automatic extraction of meaningful information from large collections of noisy data. In many real world applications, data is naturally represented in structured form. Since traditional methods in machine learning deal with vectorial information, they require an a priori form of preprocessing. Among all the learning techniques for dealing with structured data, kernel methods are recognized to have a strong theoretical background and to be effective approaches. They do not require an explicit vectorial representation of the data in terms of features, but rely on a measure of similarity between any pair of objects of a domain, the kernel function. Designing fast and good kernel functions is a challenging problem. In the case of tree structured data two issues become relevant: kernel for trees should not be sparse and should be fast to compute. The sparsity problem arises when, given a dataset and a kernel function, most structures of the dataset are completely dissimilar to one another. In those cases the classifier has too few information for making correct predictions on unseen data. In fact, it tends to produce a discriminating function behaving as the nearest neighbour rule. Sparsity is likely to arise for some standard tree kernel functions, such as the subtree and subset tree kernel, when they are applied to datasets with node labels belonging to a large domain. A second drawback of using tree kernels is the time complexity required both in learning and classification phases. Such a complexity can sometimes prevents the kernel application in scenarios involving large amount of data. This thesis proposes three contributions for resolving the above issues of kernel for trees. A first contribution aims at creating kernel functions which adapt to the statistical properties of the dataset, thus reducing its sparsity with respect to traditional tree kernel functions. Specifically, we propose to encode the input trees by an algorithm able to project the data onto a lower dimensional space with the property that similar structures are mapped similarly. By building kernel functions on the lower dimensional representation, we are able to perform inexact matchings between different inputs in the original space. A second contribution is the proposal of a novel kernel function based on the convolution kernel framework. Convolution kernel measures the similarity of two objects in terms of the similarities of their subparts. Most convolution kernels are based on counting the number of shared substructures, partially discarding information about their position in the original structure. The kernel function we propose is, instead, especially focused on this aspect. A third contribution is devoted at reducing the computational burden related to the calculation of a kernel function between a tree and a forest of trees, which is a typical operation in the classification phase and, for some algorithms, also in the learning phase. We propose a general methodology applicable to convolution kernels. Moreover, we show an instantiation of our technique when kernels such as the subtree and subset tree kernels are employed. In those cases, Direct Acyclic Graphs can be used to compactly represent shared substructures in different trees, thus reducing the computational burden and storage requirements.

Lesion detection and classification in breast cancer: evaluation of approaches based on morphological features, tracer kinetic modelling and semi-quantitative parameters in MR functional imaging (DCE-MRI)

The diagnosis, grading and classification of tumours has benefited considerably from the development of DCE-MRI which is now essential to the adequate clinical management of many tumour types due to its capability in detecting active angiogenesis. Several strategies have been proposed for DCE-MRI evaluation. Visual inspection of contrast agent concentration curves vs time is a very simple yet operator dependent procedure, therefore more objective approaches have been developed in order to facilitate comparison between studies. In so called model free approaches, descriptive or heuristic information extracted from time series raw data have been used for tissue classification. The main issue concerning these schemes is that they have not a direct interpretation in terms of physiological properties of the tissues. On the other hand, model based investigations typically involve compartmental tracer kinetic modelling and pixel-by-pixel estimation of kinetic parameters via non-linear regression applied on region of interests opportunely selected by the physician. This approach has the advantage to provide parameters directly related to the pathophysiological properties of the tissue such as vessel permeability, local regional blood flow, extraction fraction, concentration gradient between plasma and extravascular-extracellular space. Anyway, nonlinear modelling is computational demanding and the accuracy of the estimates can be affected by the signal-to-noise ratio and by the initial solutions. The principal aim of this thesis is investigate the use of semi-quantitative and quantitative parameters for segmentation and classification of breast lesion. The objectives can be subdivided as follow: describe the principal techniques to evaluate time intensity curve in DCE-MRI with focus on kinetic model proposed in literature; to evaluate the influence in parametrization choice for a classic bi-compartmental kinetic models; to evaluate the performance of a method for simultaneous tracer kinetic modelling and pixel classification; to evaluate performance of machine learning techniques training for segmentation and classification of breast lesion.

Automotive Radars:towards smarter vehicles

Radars are expected to become the main sensors in various civilian applications, especially for autonomous driving. Their success is mainly due to the availability of low cost integrated devices, equipped with compact antenna arrays, and computationally efficient signal processing techniques. This thesis focuses on the study and the development of different deterministic and learning based techniques for colocated multiple-input multiple-output (MIMO) radars. In particular, after providing an overview on the architecture of these devices, the problem of detecting and estimating multiple targets in stepped frequency continuous wave (SFCW) MIMO radar systems is investigated and different deterministic techniques solving it are illustrated. Moreover, novel solutions, based on an approximate maximum likelihood approach, are developed. The accuracy achieved by all the considered algorithms is assessed on the basis of the raw data acquired from low power wideband radar devices. The results demonstrate that the developed algorithms achieve reasonable accuracies, but at the price of different computational efforts. Another important technical problem investigated in this thesis concerns the exploitation of machine learning and deep learning techniques in the field of colocated MIMO radars. In this thesis, after providing a comprehensive overview of the machine learning and deep learning techniques currently being considered for use in MIMO radar systems, their performance in two different applications is assessed on the basis of synthetically generated and experimental datasets acquired through a commercial frequency modulated continuous wave (FMCW) MIMO radar. Finally, the application of colocated MIMO radars to autonomous driving in smart agriculture is illustrated.

Wireless solutions for the industrial internet of things

The fourth industrial revolution is paving the way for Industrial Internet of Things applications where industrial assets (e.g., robotic arms, valves, pistons) are equipped with a large number of wireless devices (i.e., microcontroller boards that embed sensors and actuators) to enable a plethora of new applications, such as analytics, diagnostics, monitoring, as well as supervisory, and safety control use-cases. Nevertheless, current wireless technologies, such as Wi-Fi, Bluetooth, and even private 5G networks, cannot fulfill all the requirements set up by the Industry 4.0 paradigm, thus opening up new 6G-oriented research trends, such as the use of THz frequencies. In light of the above, this thesis provides (i) a broad overview of the main use-cases, requirements, and key enabling wireless technologies foreseen by the fourth industrial revolution, and (ii) proposes innovative contributions, both theoretical and empirical, to enhance the performance of current and future wireless technologies at different levels of the protocol stack. In particular, at the physical layer, signal processing techniques are being exploited to analyze two multiplexing schemes, namely Affine Frequency Division Multiplexing and Orthogonal Chirp Division Multiplexing, which seem promising for high-frequency wireless communications. At the medium access layer, three protocols for intra-machine communications are proposed, where one is based on LoRa at 2.4 GHz and the others work in the THz band. Different scheduling algorithms for private industrial 5G networks are compared, and two main proposals are described, i.e., a decentralized scheme that leverages machine learning techniques to better address aperiodic traffic patterns, and a centralized contention-based design that serves a federated learning industrial application. Results are provided in terms of numerical evaluations, simulation results, and real-world experiments. Several improvements over the state-of-the-art were obtained, and the description of up-and-running testbeds demonstrates the feasibility of some of the theoretical concepts when considering a real industry plant.

Data-driven and data-oriented methods for materials science and technologies

The discovery of new materials and their functions has always been a fundamental component of technological progress. Nowadays, the quest for new materials is stronger than ever: sustainability, medicine, robotics and electronics are all key assets which depend on the ability to create specifically tailored materials. However, designing materials with desired properties is a difficult task, and the complexity of the discipline makes it difficult to identify general criteria. While scientists developed a set of best practices (often based on experience and expertise), this is still a trial-and-error process. This becomes even more complex when dealing with advanced functional materials. Their properties depend on structural and morphological features, which in turn depend on fabrication procedures and environment, and subtle alterations leads to dramatically different results. Because of this, materials modeling and design is one of the most prolific research fields. Many techniques and instruments are continuously developed to enable new possibilities, both in the experimental and computational realms. Scientists strive to enforce cutting-edge technologies in order to make progress. However, the field is strongly affected by unorganized file management, proliferation of custom data formats and storage procedures, both in experimental and computational research. Results are difficult to find, interpret and re-use, and a huge amount of time is spent interpreting and re-organizing data. This also strongly limit the application of data-driven and machine learning techniques. This work introduces possible solutions to the problems described above. Specifically, it talks about developing features for specific classes of advanced materials and use them to train machine learning models and accelerate computational predictions for molecular compounds; developing method for organizing non homogeneous materials data; automate the process of using devices simulations to train machine learning models; dealing with scattered experimental data and use them to discover new patterns.

Anticancer drug discovery using artificial intelligence: an application in pharmacological activity prediction

Hematological cancers are a heterogeneous family of diseases that can be divided into leukemias, lymphomas, and myelomas, often called “liquid tumors”. Since they cannot be surgically removable, chemotherapy represents the mainstay of their treatment. However, it still faces several challenges like drug resistance and low response rate, and the need for new anticancer agents is compelling. The drug discovery process is long-term, costly, and prone to high failure rates. With the rapid expansion of biological and chemical "big data", some computational techniques such as machine learning tools have been increasingly employed to speed up and economize the whole process. Machine learning algorithms can create complex models with the aim to determine the biological activity of compounds against several targets, based on their chemical properties. These models are defined as multi-target Quantitative Structure-Activity Relationship (mt-QSAR) and can be used to virtually screen small and large chemical libraries for the identification of new molecules with anticancer activity. The aim of my Ph.D. project was to employ machine learning techniques to build an mt-QSAR classification model for the prediction of cytotoxic drugs simultaneously active against 43 hematological cancer cell lines. For this purpose, first, I constructed a large and diversified dataset of molecules extracted from the ChEMBL database. Then, I compared the performance of different ML classification algorithms, until Random Forest was identified as the one returning the best predictions. Finally, I used different approaches to maximize the performance of the model, which achieved an accuracy of 88% by correctly classifying 93% of inactive molecules and 72% of active molecules in a validation set. This model was further applied to the virtual screening of a small dataset of molecules tested in our laboratory, where it showed 100% accuracy in correctly classifying all molecules. This result is confirmed by our previous in vitro experiments.