Continual learning for computer vision applications

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

Typed behavioural equivalences in the Pi-Calculus

In this thesis, I study the notion of program equivalences, i.e. proving that two programs can be used interchangeably without altering the overall observable behaviour. This definition is highly dependent on the contexts in which these programs can be used; does the context have exceptions, parallelism, etc... So proofs also need to be adapted according to the expressiveness of those contexts. This thesis presents on the pi-calculus – a concurrent programming language – under various typing constraints. Types allows us to impose different disciplines like forcing a sequential execution, or ensuring linearity, meaning an object can be used once. In each case, the bisimulation, a standard proof technique for the pi-calculus, needs to be adapted accordingly to obtain a suitable equivalence. We then test how using the modified bisimulations can be used to reason about a language with higher-order functions and references, which once translated into the pi-calculus satisfies the typing constraints.

Teaching informatics to novices: big ideas and the necessity of optimal guidance

This thesis reports on the two main areas of our research: introductory programming as the traditional way of accessing informatics and cultural teaching informatics through unconventional pathways. The research on introductory programming aims to overcome challenges in traditional programming education, thus increasing participation in informatics. Improving access to informatics enables individuals to pursue more and better professional opportunities and contribute to informatics advancements. We aimed to balance active, student-centered activities and provide optimal support to novices at their level. Inspired by Productive Failure and exploring the concept of notional machine, our work focused on developing Necessity Learning Design, a design to help novices tackle new programming concepts. Using this design, we implemented a learning sequence to introduce arrays and evaluated it in a real high-school context. The subsequent chapters discuss our experiences teaching CS1 in a remote-only scenario during the COVID-19 pandemic and our collaborative effort with primary school teachers to develop a learning module for teaching iteration using a visual programming environment. The research on teaching informatics principles through unconventional pathways, such as cryptography, aims to introduce informatics to a broader audience, particularly younger individuals that are less technical and professional-oriented. It emphasizes the importance of understanding informatics's cultural and scientific aspects to focus on the informatics societal value and its principles for active citizenship. After reflecting on computational thinking and inspired by the big ideas of science and informatics, we describe our hands-on approach to teaching cryptography in high school, which leverages its key scientific elements to emphasize its social aspects. Additionally, we present an activity for teaching public-key cryptography using graphs to explore fundamental concepts and methods in informatics and mathematics and their interdisciplinarity. In broadening the understanding of informatics, these research initiatives also aim to foster motivation and prime for more professional learning of informatics.

Computational methods applied to undeciphered scripts from the Aegean and Cyprus: Linear A and Cypro-Minoan

The study of ancient, undeciphered scripts presents unique challenges, that depend both on the nature of the problem and on the peculiarities of each writing system. In this thesis, I present two computational approaches that are tailored to two different tasks and writing systems. The first of these methods is aimed at the decipherment of the Linear A afraction signs, in order to discover their numerical values. This is achieved with a combination of constraint programming, ad-hoc metrics and paleographic considerations. The second main contribution of this thesis regards the creation of an unsupervised deep learning model which uses drawings of signs from ancient writing system to learn to distinguish different graphemes in the vector space. This system, which is based on techniques used in the field of computer vision, is adapted to the study of ancient writing systems by incorporating information about sequences in the model, mirroring what is often done in natural language processing. In order to develop this model, the Cypriot Greek Syllabary is used as a target, since this is a deciphered writing system. Finally, this unsupervised model is adapted to the undeciphered Cypro-Minoan and it is used to answer open questions about this script. In particular, by reconstructing multiple allographs that are not agreed upon by paleographers, it supports the idea that Cypro-Minoan is a single script and not a collection of three script like it was proposed in the literature. These results on two different tasks shows that computational methods can be applied to undeciphered scripts, despite the relatively low amount of available data, paving the way for further advancement in paleography using these methods.

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.

Towards a logical foundation of randomized computation

This dissertation investigates the relations between logic and TCS in the probabilistic setting. It is motivated by two main considerations. On the one hand, since their appearance in the 1960s-1970s, probabilistic models have become increasingly pervasive in several fast-growing areas of CS. On the other, the study and development of (deterministic) computational models has considerably benefitted from the mutual interchanges between logic and CS. Nevertheless, probabilistic computation was only marginally touched by such fruitful interactions. The goal of this thesis is precisely to (start) bring(ing) this gap, by developing logical systems corresponding to specific aspects of randomized computation and, therefore, by generalizing standard achievements to the probabilistic realm. To do so, our key ingredient is the introduction of new, measure-sensitive quantifiers associated with quantitative interpretations. The dissertation is tripartite. In the first part, we focus on the relation between logic and counting complexity classes. We show that, due to our classical counting propositional logic, it is possible to generalize to counting classes, the standard results by Cook and Meyer and Stockmeyer linking propositional logic and the polynomial hierarchy. Indeed, we show that the validity problem for counting-quantified formulae captures the corresponding level in Wagner's hierarchy. In the second part, we consider programming language theory. Type systems for randomized \lambda-calculi, also guaranteeing various forms of termination properties, were introduced in the last decades, but these are not "logically oriented" and no Curry-Howard correspondence is known for them. Following intuitions coming from counting logics, we define the first probabilistic version of the correspondence. Finally, we consider the relationship between arithmetic and computation. We present a quantitative extension of the language of arithmetic able to formalize basic results from probability theory. This language is also our starting point to define randomized bounded theories and, so, to generalize canonical results by Buss.

An architecture for network acceleration as a service in the cloud continuum

The pervasive availability of connected devices in any industrial and societal sector is pushing for an evolution of the well-established cloud computing model. The emerging paradigm of the cloud continuum embraces this decentralization trend and envisions virtualized computing resources physically located between traditional datacenters and data sources. By totally or partially executing closer to the network edge, applications can have quicker reactions to events, thus enabling advanced forms of automation and intelligence. However, these applications also induce new data-intensive workloads with low-latency constraints that require the adoption of specialized resources, such as high-performance communication options (e.g., RDMA, DPDK, XDP, etc.). Unfortunately, cloud providers still struggle to integrate these options into their infrastructures. That risks undermining the principle of generality that underlies the cloud computing scale economy by forcing developers to tailor their code to low-level APIs, non-standard programming models, and static execution environments. This thesis proposes a novel system architecture to empower cloud platforms across the whole cloud continuum with Network Acceleration as a Service (NAaaS). To provide commodity yet efficient access to acceleration, this architecture defines a layer of agnostic high-performance I/O APIs, exposed to applications and clearly separated from the heterogeneous protocols, interfaces, and hardware devices that implement it. A novel system component embodies this decoupling by offering a set of agnostic OS features to applications: memory management for zero-copy transfers, asynchronous I/O processing, and efficient packet scheduling. This thesis also explores the design space of the possible implementations of this architecture by proposing two reference middleware systems and by adopting them to support interactive use cases in the cloud continuum: a serverless platform and an Industry 4.0 scenario. A detailed discussion and a thorough performance evaluation demonstrate that the proposed architecture is suitable to enable the easy-to-use, flexible integration of modern network acceleration into next-generation cloud platforms.

Machine Learning in Automatic Tabulation for Constraint Model Reformulation

Combinatorial decision and optimization problems belong to numerous applications, such as logistics and scheduling, and can be solved with various approaches. Boolean Satisfiability and Constraint Programming solvers are some of the most used ones and their performance is significantly influenced by the model chosen to represent a given problem. This has led to the study of model reformulation methods, one of which is tabulation, that consists in rewriting the expression of a constraint in terms of a table constraint. To apply it, one should identify which constraints can help and which can hinder the solving process. So far this has been performed by hand, for example in MiniZinc, or automatically with manually designed heuristics, in Savile Row. Though, it has been shown that the performances of these heuristics differ across problems and solvers, in some cases helping and in others hindering the solving procedure. However, recent works in the field of combinatorial optimization have shown that Machine Learning (ML) can be increasingly useful in the model reformulation steps. This thesis aims to design a ML approach to identify the instances for which Savile Row’s heuristics should be activated. Additionally, it is possible that the heuristics miss some good tabulation opportunities, so we perform an exploratory analysis for the creation of a ML classifier able to predict whether or not a constraint should be tabulated. The results reached towards the first goal show that a random forest classifier leads to an increase in the performances of 4 different solvers. The experimental results in the second task show that a ML approach could improve the performance of a solver for some problem classes.

The Use of Latent Semantic Indexing to Mitigate OCR Effects of Related Document Images

Due to both the widespread and multipurpose use of document images and the current availability of a high number of document images repositories, robust information retrieval mechanisms and systems have been increasingly demanded. This paper presents an approach to support the automatic generation of relationships among document images by exploiting Latent Semantic Indexing (LSI) and Optical Character Recognition (OCR). We developed the LinkDI (Linking of Document Images) service, which extracts and indexes document images content, computes its latent semantics, and defines relationships among images as hyperlinks. LinkDI was experimented with document images repositories, and its performance was evaluated by comparing the quality of the relationships created among textual documents as well as among their respective document images. Considering those same document images, we ran further experiments in order to compare the performance of LinkDI when it exploits or not the LSI technique. Experimental results showed that LSI can mitigate the effects of usual OCR misrecognition, which reinforces the feasibility of LinkDI relating OCR output with high degradation.

Biologically Plausible Connectionist Prediction of Natural Language Thematic Relations

In Natural Language Processing (NLP) symbolic systems, several linguistic phenomena, for instance, the thematic role relationships between sentence constituents, such as AGENT, PATIENT, and LOCATION, can be accounted for by the employment of a rule-based grammar. Another approach to NLP concerns the use of the connectionist model, which has the benefits of learning, generalization and fault tolerance, among others. A third option merges the two previous approaches into a hybrid one: a symbolic thematic theory is used to supply the connectionist network with initial knowledge. Inspired on neuroscience, it is proposed a symbolic-connectionist hybrid system called BIO theta PRED (BIOlogically plausible thematic (theta) symbolic-connectionist PREDictor), designed to reveal the thematic grid assigned to a sentence. Its connectionist architecture comprises, as input, a featural representation of the words (based on the verb/noun WordNet classification and on the classical semantic microfeature representation), and, as output, the thematic grid assigned to the sentence. BIO theta PRED is designed to ""predict"" thematic (semantic) roles assigned to words in a sentence context, employing biologically inspired training algorithm and architecture, and adopting a psycholinguistic view of thematic theory.

Systematic Management of Variability in UML-based Software Product Lines

This paper presents SMarty, a variability management approach for UML-based software product lines (PL). SMarty is supported by a UML profile, the SMartyProfile, and a process for managing variabilities, the SMartyProcess. SMartyProfile aims at representing variabilities, variation points, and variants in UML models by applying a set of stereotypes. SMartyProcess consists of a set of activities that is systematically executed to trace, identify, and control variabilities in a PL based on SMarty. It also identifies variability implementation mechanisms and analyzes specific product configurations. In addition, a more comprehensive application of SMarty is presented using SEI's Arcade Game Maker PL. An evaluation of SMarty and related work are discussed.

Gene Expression Complex Networks: Synthesis, Identification, and Analysis

Thanks to recent advances in molecular biology, allied to an ever increasing amount of experimental data, the functional state of thousands of genes can now be extracted simultaneously by using methods such as cDNA microarrays and RNA-Seq. Particularly important related investigations are the modeling and identification of gene regulatory networks from expression data sets. Such a knowledge is fundamental for many applications, such as disease treatment, therapeutic intervention strategies and drugs design, as well as for planning high-throughput new experiments. Methods have been developed for gene networks modeling and identification from expression profiles. However, an important open problem regards how to validate such approaches and its results. This work presents an objective approach for validation of gene network modeling and identification which comprises the following three main aspects: (1) Artificial Gene Networks (AGNs) model generation through theoretical models of complex networks, which is used to simulate temporal expression data; (2) a computational method for gene network identification from the simulated data, which is founded on a feature selection approach where a target gene is fixed and the expression profile is observed for all other genes in order to identify a relevant subset of predictors; and (3) validation of the identified AGN-based network through comparison with the original network. The proposed framework allows several types of AGNs to be generated and used in order to simulate temporal expression data. The results of the network identification method can then be compared to the original network in order to estimate its properties and accuracy. Some of the most important theoretical models of complex networks have been assessed: the uniformly-random Erdos-Renyi (ER), the small-world Watts-Strogatz (WS), the scale-free Barabasi-Albert (BA), and geographical networks (GG). The experimental results indicate that the inference method was sensitive to average degree k variation, decreasing its network recovery rate with the increase of k. The signal size was important for the inference method to get better accuracy in the network identification rate, presenting very good results with small expression profiles. However, the adopted inference method was not sensible to recognize distinct structures of interaction among genes, presenting a similar behavior when applied to different network topologies. In summary, the proposed framework, though simple, was adequate for the validation of the inferred networks by identifying some properties of the evaluated method, which can be extended to other inference methods.

On the k-restricted structure ratio in planar and outerplanar graphs

A planar k-restricted structure is a simple graph whose blocks are planar and each has at most k vertices. Planar k-restricted structures are used by approximation algorithms for Maximum Weight Planar Subgraph, which motivates this work. The planar k-restricted ratio is the infimum, over simple planar graphs H, of the ratio of the number of edges in a maximum k-restricted structure subgraph of H to the number edges of H. We prove that, as k tends to infinity, the planar k-restricted ratio tends to 1/2. The same result holds for the weighted version. Our results are based on analyzing the analogous ratios for outerplanar and weighted outerplanar graphs. Here both ratios tend to 1 as k goes to infinity, and we provide good estimates of the rates of convergence, showing that they differ in the weighted from the unweighted case.

On the asymptotic enumeration of accessible automata

We simplify the known formula for the asymptotic estimate of the number of deterministic and accessible automata with n states over a k-letter alphabet. The proof relies on the theory of Lagrange inversion applied in the context of generalized binomial series.

Kernel machines for epilepsy diagnosis via EEG signal classification: A comparative study

Objective: We carry out a systematic assessment on a suite of kernel-based learning machines while coping with the task of epilepsy diagnosis through automatic electroencephalogram (EEG) signal classification. Methods and materials: The kernel machines investigated include the standard support vector machine (SVM), the least squares SVM, the Lagrangian SVM, the smooth SVM, the proximal SVM, and the relevance vector machine. An extensive series of experiments was conducted on publicly available data, whose clinical EEG recordings were obtained from five normal subjects and five epileptic patients. The performance levels delivered by the different kernel machines are contrasted in terms of the criteria of predictive accuracy, sensitivity to the kernel function/parameter value, and sensitivity to the type of features extracted from the signal. For this purpose, 26 values for the kernel parameter (radius) of two well-known kernel functions (namely. Gaussian and exponential radial basis functions) were considered as well as 21 types of features extracted from the EEG signal, including statistical values derived from the discrete wavelet transform, Lyapunov exponents, and combinations thereof. Results: We first quantitatively assess the impact of the choice of the wavelet basis on the quality of the features extracted. Four wavelet basis functions were considered in this study. Then, we provide the average accuracy (i.e., cross-validation error) values delivered by 252 kernel machine configurations; in particular, 40%/35% of the best-calibrated models of the standard and least squares SVMs reached 100% accuracy rate for the two kernel functions considered. Moreover, we show the sensitivity profiles exhibited by a large sample of the configurations whereby one can visually inspect their levels of sensitiveness to the type of feature and to the kernel function/parameter value. Conclusions: Overall, the results evidence that all kernel machines are competitive in terms of accuracy, with the standard and least squares SVMs prevailing more consistently. Moreover, the choice of the kernel function and parameter value as well as the choice of the feature extractor are critical decisions to be taken, albeit the choice of the wavelet family seems not to be so relevant. Also, the statistical values calculated over the Lyapunov exponents were good sources of signal representation, but not as informative as their wavelet counterparts. Finally, a typical sensitivity profile has emerged among all types of machines, involving some regions of stability separated by zones of sharp variation, with some kernel parameter values clearly associated with better accuracy rates (zones of optimality). (C) 2011 Elsevier B.V. All rights reserved.