Tensor-Train decomposition for image classification problems

In these last years a great effort has been put in the development of new techniques for automatic object classification, also due to the consequences in many applications such as medical imaging or driverless cars. To this end, several mathematical models have been developed from logistic regression to neural networks. A crucial aspect of these so called classification algorithms is the use of algebraic tools to represent and approximate the input data. In this thesis, we examine two different models for image classification based on a particular tensor decomposition named Tensor-Train (TT) decomposition. The use of tensor approaches preserves the multidimensional structure of the data and the neighboring relations among pixels. Furthermore the Tensor-Train, differently from other tensor decompositions, does not suffer from the curse of dimensionality making it an extremely powerful strategy when dealing with high-dimensional data. It also allows data compression when combined with truncation strategies that reduce memory requirements without spoiling classification performance. The first model we propose is based on a direct decomposition of the database by means of the TT decomposition to find basis vectors used to classify a new object. The second model is a tensor dictionary learning model, based on the TT decomposition where the terms of the decomposition are estimated using a proximal alternating linearized minimization algorithm with a spectral stepsize.

Rich media content adaptation in e-learning systems

The wide use of e-technologies represents a great opportunity for underserved segments of the population, especially with the aim of reintegrating excluded individuals back into society through education. This is particularly true for people with different types of disabilities who may have difficulties while attending traditional on-site learning programs that are typically based on printed learning resources. The creation and provision of accessible e-learning contents may therefore become a key factor in enabling people with different access needs to enjoy quality learning experiences and services. Another e-learning challenge is represented by m-learning (which stands for mobile learning), which is emerging as a consequence of mobile terminals diffusion and provides the opportunity to browse didactical materials everywhere, outside places that are traditionally devoted to education. Both such situations share the need to access materials in limited conditions and collide with the growing use of rich media in didactical contents, which are designed to be enjoyed without any restriction. Nowadays, Web-based teaching makes great use of multimedia technologies, ranging from Flash animations to prerecorded video-lectures. Rich media in e-learning can offer significant potential in enhancing the learning environment, through helping to increase access to education, enhance the learning experience and support multiple learning styles. Moreover, they can often be used to improve the structure of Web-based courses. These highly variegated and structured contents may significantly improve the quality and the effectiveness of educational activities for learners. For example, rich media contents allow us to describe complex concepts and process flows. Audio and video elements may be utilized to add a “human touch” to distance-learning courses. Finally, real lectures may be recorded and distributed to integrate or enrich on line materials. A confirmation of the advantages of these approaches can be seen in the exponential growth of video-lecture availability on the net, due to the ease of recording and delivering activities which take place in a traditional classroom. Furthermore, the wide use of assistive technologies for learners with disabilities injects new life into e-learning systems. E-learning allows distance and flexible educational activities, thus helping disabled learners to access resources which would otherwise present significant barriers for them. For instance, students with visual impairments have difficulties in reading traditional visual materials, deaf learners have trouble in following traditional (spoken) lectures, people with motion disabilities have problems in attending on-site programs. As already mentioned, the use of wireless technologies and pervasive computing may really enhance the educational learner experience by offering mobile e-learning services that can be accessed by handheld devices. This new paradigm of educational content distribution maximizes the benefits for learners since it enables users to overcome constraints imposed by the surrounding environment. While certainly helpful for users without disabilities, we believe that the use of newmobile technologies may also become a fundamental tool for impaired learners, since it frees them from sitting in front of a PC. In this way, educational activities can be enjoyed by all the users, without hindrance, thus increasing the social inclusion of non-typical learners. While the provision of fully accessible and portable video-lectures may be extremely useful for students, it is widely recognized that structuring and managing rich media contents for mobile learning services are complex and expensive tasks. Indeed, major difficulties originate from the basic need to provide a textual equivalent for each media resource composing a rich media Learning Object (LO). Moreover, tests need to be carried out to establish whether a given LO is fully accessible to all kinds of learners. Unfortunately, both these tasks are truly time-consuming processes, depending on the type of contents the teacher is writing and on the authoring tool he/she is using. Due to these difficulties, online LOs are often distributed as partially accessible or totally inaccessible content. Bearing this in mind, this thesis aims to discuss the key issues of a system we have developed to deliver accessible, customized or nomadic learning experiences to learners with different access needs and skills. To reduce the risk of excluding users with particular access capabilities, our system exploits Learning Objects (LOs) which are dynamically adapted and transcoded based on the specific needs of non-typical users and on the barriers that they can encounter in the environment. The basic idea is to dynamically adapt contents, by selecting them from a set of media resources packaged in SCORM-compliant LOs and stored in a self-adapting format. The system schedules and orchestrates a set of transcoding processes based on specific learner needs, so as to produce a customized LO that can be fully enjoyed by any (impaired or mobile) student.

Learning Methods and Algorithms for Semantic Text Classification across Multiple Domains

Information is nowadays a key resource: machine learning and data mining techniques have been developed to extract high-level information from great amounts of data. As most data comes in form of unstructured text in natural languages, research on text mining is currently very active and dealing with practical problems. Among these, text categorization deals with the automatic organization of large quantities of documents in priorly defined taxonomies of topic categories, possibly arranged in large hierarchies. In commonly proposed machine learning approaches, classifiers are automatically trained from pre-labeled documents: they can perform very accurate classification, but often require a consistent training set and notable computational effort. Methods for cross-domain text categorization have been proposed, allowing to leverage a set of labeled documents of one domain to classify those of another one. Most methods use advanced statistical techniques, usually involving tuning of parameters. A first contribution presented here is a method based on nearest centroid classification, where profiles of categories are generated from the known domain and then iteratively adapted to the unknown one. Despite being conceptually simple and having easily tuned parameters, this method achieves state-of-the-art accuracy in most benchmark datasets with fast running times. A second, deeper contribution involves the design of a domain-independent model to distinguish the degree and type of relatedness between arbitrary documents and topics, inferred from the different types of semantic relationships between respective representative words, identified by specific search algorithms. The application of this model is tested on both flat and hierarchical text categorization, where it potentially allows the efficient addition of new categories during classification. Results show that classification accuracy still requires improvements, but models generated from one domain are shown to be effectively able to be reused in a different one.

Learning representations for graph-structured socio-technical systems

The recent widespread use of social media platforms and web services has led to a vast amount of behavioral data that can be used to model socio-technical systems. A significant part of this data can be represented as graphs or networks, which have become the prevalent mathematical framework for studying the structure and the dynamics of complex interacting systems. However, analyzing and understanding these data presents new challenges due to their increasing complexity and diversity. For instance, the characterization of real-world networks includes the need of accounting for their temporal dimension, together with incorporating higher-order interactions beyond the traditional pairwise formalism. The ongoing growth of AI has led to the integration of traditional graph mining techniques with representation learning and low-dimensional embeddings of networks to address current challenges. These methods capture the underlying similarities and geometry of graph-shaped data, generating latent representations that enable the resolution of various tasks, such as link prediction, node classification, and graph clustering. As these techniques gain popularity, there is even a growing concern about their responsible use. In particular, there has been an increased emphasis on addressing the limitations of interpretability in graph representation learning. This thesis contributes to the advancement of knowledge in the field of graph representation learning and has potential applications in a wide range of complex systems domains. We initially focus on forecasting problems related to face-to-face contact networks with time-varying graph embeddings. Then, we study hyperedge prediction and reconstruction with simplicial complex embeddings. Finally, we analyze the problem of interpreting latent dimensions in node embeddings for graphs. The proposed models are extensively evaluated in multiple experimental settings and the results demonstrate their effectiveness and reliability, achieving state-of-the-art performances and providing valuable insights into the properties of the learned representations.

La costruzione razionale della casa. Scritti e progetti di Giuseppe Pagano

The rational construction of the house. The writings and projects of Giuseppe Pagano Description, themes and research objectives The research aims at analysing the architecture of Giuseppe Pagano, which focuses on the theme of dwelling, through the reading of 3 of his house projects. On the one hand, these projects represent “minor” works not thoroughly known by Pagano’s contemporary critics; on the other they emphasise a particular methodological approach, which serves the author to explore a theme closely linked to his theoretical thought. The house project is a key to Pagano’s research, given its ties to the socio-cultural and political conditions in which the architect was working, so that it becomes a mirror of one of his specific and theoretical path, always in a state of becoming. Pagano understands architecture as a “servant of the human being”, subject to a “utilitarian slavery” since it is a clear, essential and “modest” answer to specific human needs, free from aprioristic aesthetic and formal choices. It is a rational architecture in sensu stricto; it constitutes a perfect synthesis between cause and effect and between function and form. The house needs to accommodate these principles because it is closely intertwined with human needs and intimately linked to a specific place, climatic conditions and technical and economical possibilities. Besides, differently from his public and common masterpieces such as the Palazzo Gualino, the Istituto di Fisica and the Università Commerciale Bocconi, the house projects are representative of a precise project will, which is expressed in a more authentic way, partially freed from political influences and dogmatic preoccupations and, therefore, far from the attempt to research a specific expressive language. I believe that the house project better represents that “ingenuity”, freshness and “sincerity” that Pagano identifies with the minor architecture, thereby revealing a more authentic expression of his understanding of a project. Therefore, the thesis, by tracing the theoretical research of Pagano through the analysis of some of his designed and built works, attempts to identify a specific methodological approach to Pagano’s project, which, developed through time, achieves a certain clarity in the 1930s. In fact, this methodological approach becomes more evident in his last projects, mainly regarding the house and the urban space. These reflect the attempt to respond to the new social needs and, at the same time, they also are an expression of a freer idea of built architecture, closely linked with the place and with the human being who dwells it. The three chosen projects (Villa Colli, La Casa a struttura d’acciaio and Villa Caraccio) make Pagano facing different places, different customers and different economic and technical conditions, which, given the author’s biography, correspond to important historical and political conditions. This is the reason why the projects become apparently distant works, both linguistically and conceptually, to the point that one can define them as ”eclectic”. However, I argue that this eclecticism is actually an added value to the architectural work of Pagano, steaming from the use of a method which, having as a basis the postulate of a rational architecture as essence and logic of building, finds specific variations depending on the multiple variables to be addressed by the project. This is the methodological heritage that Pagano learns from the tradition, especially that of the rural residential architecture, defined by Pagano as a “dictionary of the building logic of man”, as an “a-stylistic background”. For Pagano this traditional architecture is a clear expression of the relationships between a theme and its development, an architectural “fact” that is resolved with purely technical and utilitarian aims and with a spontaneous development far from any aprioristic theoretical principle. Architecture, therefore, cannot be an invention for Pagano and the personal contribution of each architect has to consider his/her close relationship with the specific historical context, place and new building methods. These are basic principles in the methodological approach that drives a great deal of his research and that also permits his thought to be modern. I argue that both ongoing and new collaborations with younger protagonists of the culture and architecture of the period are significant for the development of his methodology. These encounters represent the will to spread his own understanding of the “new architecture” as well as a way of self-renewal by confronting the self with new themes and realities and by learning from his collaborators. Thesis’ outline The thesis is divided in two principal parts, each articulated in four chapters attempting to offer a new reading of the theory and work of Pagano by emphasising the central themes of the research. The first chapter is an introduction to the thesis and to the theme of the rational house, as understood and developed in its typological and technical aspects by Pagano and by other protagonists of the Italian rationalism of the 1930s. Here the attention is on two different aspects defining, according to Pagano, the house project: on the one hand, the typological renewal, aimed at defining a “standard form” as a clear and essential answer to certain needs and variables of the project leading to different formal expressions. On the other, it focuses on the building, understood as a technique to “produce” architecture, where new technologies and new materials are not merely tools but also essential elements of the architectural work. In this way the villa becomes different from the theme of the common house or from that of the minimalist house, by using rules in the choice of material and in the techniques that are every time different depending on the theme under exploration and on the contingency of place. It is also visible the rigorous rationalism that distinguishes the author's appropriation of certain themes of rural architecture. The pages of “Casabella” and the events of the contemporary Triennali form the preliminary material for the writing of this chapter given that they are primary sources to individuate projects and writings produced by Pagano and contemporary architects on this theme. These writings and projects, when compared, reconstruct the evolution of the idea of the rational house and, specifically, of the personal research of Pagano. The second part regards the reading of three of Pagano’s projects of houses as a built verification of his theories. This section constitutes the central part of the thesis since it is aimed at detecting a specific methodological approach showing a theoretical and ideological evolution expressed in the vast edited literature. The three projects that have been chosen explore the theme of the house, looking at various research themes that the author proposes and that find continuity in the affirmation of a specific rationalism, focussed on concepts such as essentiality, utility, functionality and building honesty. These concepts guide the thought and the activities of Pagano, also reflecting a social and cultural period. The projects span from the theme of the villa moderna, Villa Colli, which, inspired by the architecture of North Europe, anticipates a specific rationalism of Pagano based on rigour, simplicity and essentiality, to the theme of the common house, Casa a struttura d’acciaio, la casa del domani, which ponders on the definition of new living spaces and, moreover, on new concepts of standardisation, economical efficiency and new materials responding to the changing needs of the modern society. Finally, the third project returns to the theme of the, Villa Caraccio, revisiting it with new perspectives. These perspectives find in the solution of the open plant, in the openness to nature and landscape and in the revisiting of materials and local building systems that idea of the freed house, which express clearly a new theoretical thought. Methodology It needs to be noted that due to the lack of an official Archive of Pagano’s work, the analysis of his work has been difficult and this explains the necessity to read the articles and the drawings published in the pages of «Casabella» and «Domus». As for the projects of Villa Colli and Casa a struttura d’acciaio, parts of the original drawings have been consulted. These drawings are not published and are kept in private archives of the collaborators of Pagano. The consultation of these documents has permitted the analysis of the cited works, which have been subject to a more complete reading following the different proposed solutions, which have permitted to understand the project path. The projects are analysed thought the method of comparison and critical reading which, specifically, means graphical elaborations and analytical schemes, mostly reconstructed on the basis of original projects but, where possible, also on a photographic investigation. The focus is on the project theme which, beginning with a specific living (dwelling) typology, finds variations because of the historico-political context in which Pagano is embedded and which partially shapes his research and theoretical thought, then translated in the built work. The analysis of the work follows, beginning, where possible, from a reconstruction of the evolution of the project as elaborated on the basis of the original documents and ending on an analysis of the constructive principles and composition. This second phase employs a methodology proposed by Pagano in his article Piante di ville, which, as expected, focuses on the plant as essential tool to identify the “true practical and poetic qualities of the construction”(Pagano, «Costruzioni-Casabella», 1940, p. 2). The reading of the project is integrated with the constructive analyses related to the technical aspects of the house which, in the case of Casa a struttura d’acciaio, play an important role in the project, while in Villa Colli and in Villa Caraccio are principally linked to the choice of materials for the construction of the different architectural elements. These are nonetheless key factors in the composition of the work. Future work could extend this reading to other house projects to deepen the research that could be completed with the consultation of Archival materials, which are missing at present. Finally, in the appendix I present a critical selection of the Pagano’s writings, which recall the themes discussed and embodied by the three projects. The texts have been selected among the articles published in Casabella and in other journals, completing the reading of the project work which cannot be detached from his theoretical thought. Moving from theory to project, we follow a path that brings us to define and deepen the central theme of the thesis: rational building as the principal feature of the architectural research of Pagano, which is paraphrased in multiple ways in his designed and built works.

From core rigidities to dynamic capabilities: the role of external knowledge. A multiple case study

This study focuses on the processes of change that firms undertake to overcome conditions of organizational rigidity and develop new dynamic capabilities, thanks to the contribution of external knowledge. When external contingencies highlight firms’ core rigidities, external actors can intervene in change projects, providing new competences to firms’ managers. Knowledge transfer and organizational learning processes can lead to the development of new dynamic capabilities. Existing literature does not completely explain how these processes develop and how external knowledge providers, as management consultants, influence them. Dynamic capabilities literature has become very rich in the last years; however, the models that explain how dynamic capabilities evolve are not particularly investigated. Adopting a qualitative approach, this research proposes four relevant case studies in which external actors introduce new knowledge within organizations, activating processes of change. Each case study consists of a management consulting project. Data are collected through in-depth interviews with consultants and managers. A large amount of documents supports evidences from interviews. A narrative approach is adopted to account for change processes and a synthetic approach is proposed to compare case studies along relevant dimensions. This study presents a model of capabilities evolution, supported by empirical evidence, to explain how external knowledge intervenes in capabilities evolution processes: first, external actors solve gaps between environmental demands and firms’ capabilities, changing organizational structures and routines; second, a knowledge transfer between consultants and managers leads to the creation of new ordinary capabilities; third, managers can develop new dynamic capabilities through a deliberate learning process that internalizes new tacit knowledge from consultants. After the end of the consulting project, two elements can influence the deliberate learning process: new external contingencies and changes in the perceptions about external actors.

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.

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.

Unsupervised reinforcement learning via state entropy maximization

Reinforcement Learning (RL) provides a powerful framework to address sequential decision-making problems in which the transition dynamics is unknown or too complex to be represented. The RL approach is based on speculating what is the best decision to make given sample estimates obtained from previous interactions, a recipe that led to several breakthroughs in various domains, ranging from game playing to robotics. Despite their success, current RL methods hardly generalize from one task to another, and achieving the kind of generalization obtained through unsupervised pre-training in non-sequential problems seems unthinkable. Unsupervised RL has recently emerged as a way to improve generalization of RL methods. Just as its non-sequential counterpart, the unsupervised RL framework comprises two phases: An unsupervised pre-training phase, in which the agent interacts with the environment without external feedback, and a supervised fine-tuning phase, in which the agent aims to efficiently solve a task in the same environment by exploiting the knowledge acquired during pre-training. In this thesis, we study unsupervised RL via state entropy maximization, in which the agent makes use of the unsupervised interactions to pre-train a policy that maximizes the entropy of its induced state distribution. First, we provide a theoretical characterization of the learning problem by considering a convex RL formulation that subsumes state entropy maximization. Our analysis shows that maximizing the state entropy in finite trials is inherently harder than RL. Then, we study the state entropy maximization problem from an optimization perspective. Especially, we show that the primal formulation of the corresponding optimization problem can be (approximately) addressed through tractable linear programs. Finally, we provide the first practical methodologies for state entropy maximization in complex domains, both when the pre-training takes place in a single environment as well as multiple environments.

Real-time anomaly detection of gamma-ray bursts for the Cherenkov Telescope Array using deep learning

The Cherenkov Telescope Array (CTA) will be the next-generation ground-based observatory to study the universe in the very-high-energy domain. The observatory will rely on a Science Alert Generation (SAG) system to analyze the real-time data from the telescopes and generate science alerts. The SAG system will play a crucial role in the search and follow-up of transients from external alerts, enabling multi-wavelength and multi-messenger collaborations. It will maximize the potential for the detection of the rarest phenomena, such as gamma-ray bursts (GRBs), which are the science case for this study. This study presents an anomaly detection method based on deep learning for detecting gamma-ray burst events in real-time. The performance of the proposed method is evaluated and compared against the Li&Ma standard technique in two use cases of serendipitous discoveries and follow-up observations, using short exposure times. The method shows promising results in detecting GRBs and is flexible enough to allow real-time search for transient events on multiple time scales. The method does not assume background nor source models and doe not require a minimum number of photon counts to perform analysis, making it well-suited for real-time analysis. Future improvements involve further tests, relaxing some of the assumptions made in this study as well as post-trials correction of the detection significance. Moreover, the ability to detect other transient classes in different scenarios must be investigated for completeness. The system can be integrated within the SAG system of CTA and deployed on the onsite computing clusters. This would provide valuable insights into the method's performance in a real-world setting and be another valuable tool for discovering new transient events in real-time. Overall, this study makes a significant contribution to the field of astrophysics by demonstrating the effectiveness of deep learning-based anomaly detection techniques for real-time source detection in gamma-ray astronomy.

Development of machine learning methods for multi-modal biomarkers detection and integration

In medicine, innovation depends on a better knowledge of the human body mechanism, which represents a complex system of multi-scale constituents. Unraveling the complexity underneath diseases proves to be challenging. A deep understanding of the inner workings comes with dealing with many heterogeneous information. Exploring the molecular status and the organization of genes, proteins, metabolites provides insights on what is driving a disease, from aggressiveness to curability. Molecular constituents, however, are only the building blocks of the human body and cannot currently tell the whole story of diseases. This is why nowadays attention is growing towards the contemporary exploitation of multi-scale information. Holistic methods are then drawing interest to address the problem of integrating heterogeneous data. The heterogeneity may derive from the diversity across data types and from the diversity within diseases. Here, four studies conducted data integration using customly designed workflows that implement novel methods and views to tackle the heterogeneous characterization of diseases. The first study devoted to determine shared gene regulatory signatures for onco-hematology and it showed partial co-regulation across blood-related diseases. The second study focused on Acute Myeloid Leukemia and refined the unsupervised integration of genomic alterations, which turned out to better resemble clinical practice. In the third study, network integration for artherosclerosis demonstrated, as a proof of concept, the impact of network intelligibility when it comes to model heterogeneous data, which showed to accelerate the identification of new potential pharmaceutical targets. Lastly, the fourth study introduced a new method to integrate multiple data types in a unique latent heterogeneous-representation that facilitated the selection of important data types to predict the tumour stage of invasive ductal carcinoma. The results of these four studies laid the groundwork to ease the detection of new biomarkers ultimately beneficial to medical practice and to the ever-growing field of Personalized Medicine.