Learning to recognize faces incrementally

[EN]Most face recognition systems are based on some form of batch learning. Online face recognition is not only more practical, it is also much more biologically plausible. Typical batch learners aim at minimizing both training error and (a measure of) hypothesis complexity. We show that the same minimization can be done incrementally as long as some form of ”scaffolding” is applied throughout the learning process. Scaffolding means: make the system learn from samples that are neither too easy nor too difficult at each step. We note that such learning behavior is also biologically plausible. Experiments using large sequences of facial images support the theoretical claims. The proposed method compares well with other, numerical calculus-based online learners.

Data Science Seminar: Generic Big Data Processing for Advancing Situation Awareness and Decision-Support

The generation of heterogeneous big data sources with ever increasing volumes, velocities and veracities over the he last few years has inspired the data science and research community to address the challenge of extracting knowledge form big data. Such a wealth of generated data across the board can be intelligently exploited to advance our knowledge about our environment, public health, critical infrastructure and security. In recent years we have developed generic approaches to process such big data at multiple levels for advancing decision-support. It specifically concerns data processing with semantic harmonisation, low level fusion, analytics, knowledge modelling with high level fusion and reasoning. Such approaches will be introduced and presented in context of the TRIDEC project results on critical oil and gas industry drilling operations and also the ongoing large eVacuate project on critical crowd behaviour detection in confined spaces.

Data Science Seminar: The data science revolution in Physics and Astronomy

Abstract Heading into the 2020s, Physics and Astronomy are undergoing experimental revolutions that will reshape our picture of the fabric of the Universe. The Large Hadron Collider (LHC), the largest particle physics project in the world, produces 30 petabytes of data annually that need to be sifted through, analysed, and modelled. In astrophysics, the Large Synoptic Survey Telescope (LSST) will be taking a high-resolution image of the full sky every 3 days, leading to data rates of 30 terabytes per night over ten years. These experiments endeavour to answer the question why 96% of the content of the universe currently elude our physical understanding. Both the LHC and LSST share the 5-dimensional nature of their data, with position, energy and time being the fundamental axes. This talk will present an overview of the experiments and data that is gathered, and outlines the challenges in extracting information. Common strategies employed are very similar to industrial data! Science problems (e.g., data filtering, machine learning, statistical interpretation) and provide a seed for exchange of knowledge between academia and industry. Speaker Biography Professor Mark Sullivan Mark Sullivan is a Professor of Astrophysics in the Department of Physics and Astronomy. Mark completed his PhD at Cambridge, and following postdoctoral study in Durham, Toronto and Oxford, now leads a research group at Southampton studying dark energy using exploding stars called "type Ia supernovae". Mark has many years' experience of research that involves repeatedly imaging the night sky to track the arrival of transient objects, involving significant challenges in data handling, processing, classification and analysis.

Common Environment for Undergraduate Computer Programming

We present an Integrated Environment suitable for learning and teaching computer programming which is designed for both students of specialised Computer Science courses, and also non-specialist students such as those following Liberal Arts. The environment is rich enough to allow exploration of concepts from robotics, artificial intelligence, social science, and philosophy as well as the specialist areas of operating systems and the various computer programming paradigms.

The design and Evaluation of 'CAP Tools' - A computer aided parallelisation tool-kit

Abstract not available

Anomaly Detection in Noisy Images

Finding rare events in multidimensional data is an important detection problem that has applications in many fields, such as risk estimation in insurance industry, finance, flood prediction, medical diagnosis, quality assurance, security, or safety in transportation. The occurrence of such anomalies is so infrequent that there is usually not enough training data to learn an accurate statistical model of the anomaly class. In some cases, such events may have never been observed, so the only information that is available is a set of normal samples and an assumed pairwise similarity function. Such metric may only be known up to a certain number of unspecified parameters, which would either need to be learned from training data, or fixed by a domain expert. Sometimes, the anomalous condition may be formulated algebraically, such as a measure exceeding a predefined threshold, but nuisance variables may complicate the estimation of such a measure. Change detection methods used in time series analysis are not easily extendable to the multidimensional case, where discontinuities are not localized to a single point. On the other hand, in higher dimensions, data exhibits more complex interdependencies, and there is redundancy that could be exploited to adaptively model the normal data. In the first part of this dissertation, we review the theoretical framework for anomaly detection in images and previous anomaly detection work done in the context of crack detection and detection of anomalous components in railway tracks. In the second part, we propose new anomaly detection algorithms. The fact that curvilinear discontinuities in images are sparse with respect to the frame of shearlets, allows us to pose this anomaly detection problem as basis pursuit optimization. Therefore, we pose the problem of detecting curvilinear anomalies in noisy textured images as a blind source separation problem under sparsity constraints, and propose an iterative shrinkage algorithm to solve it. Taking advantage of the parallel nature of this algorithm, we describe how this method can be accelerated using graphical processing units (GPU). Then, we propose a new method for finding defective components on railway tracks using cameras mounted on a train. We describe how to extract features and use a combination of classifiers to solve this problem. Then, we scale anomaly detection to bigger datasets with complex interdependencies. We show that the anomaly detection problem naturally fits in the multitask learning framework. The first task consists of learning a compact representation of the good samples, while the second task consists of learning the anomaly detector. Using deep convolutional neural networks, we show that it is possible to train a deep model with a limited number of anomalous examples. In sequential detection problems, the presence of time-variant nuisance parameters affect the detection performance. In the last part of this dissertation, we present a method for adaptively estimating the threshold of sequential detectors using Extreme Value Theory on a Bayesian framework. Finally, conclusions on the results obtained are provided, followed by a discussion of possible future work.

When is bigger better? The effects of group size on the evolution of helping behaviours.

Understanding the evolution of sociality in humans and other species requires understanding how selection on social behaviour varies with group size. However, the effects of group size are frequently obscured in the theoretical literature, which often makes assumptions that are at odds with empirical findings. In particular, mechanisms are suggested as supporting large-scale cooperation when they would in fact rapidly become ineffective with increasing group size. Here we review the literature on the evolution of helping behaviours (cooperation and altruism), and frame it using a simple synthetic model that allows us to delineate how the three main components of the selection pressure on helping must vary with increasing group size. The first component is the marginal benefit of helping to group members, which determines both direct fitness benefits to the actor and indirect fitness benefits to recipients. While this is often assumed to be independent of group size, marginal benefits are in practice likely to be maximal at intermediate group sizes for many types of collective action problems, and will eventually become very small in large groups due to the law of decreasing returns. The second component is the response of social partners on the past play of an actor, which underlies conditional behaviour under repeated social interactions. We argue that under realistic conditions on the transmission of information in a population, this response on past play decreases rapidly with increasing group size so that reciprocity alone (whether direct, indirect, or generalised) cannot sustain cooperation in very large groups. The final component is the relatedness between actor and recipient, which, according to the rules of inheritance, again decreases rapidly with increasing group size. These results explain why helping behaviours in very large social groups are limited to cases where the number of reproducing individuals is small, as in social insects, or where there are social institutions that can promote (possibly through sanctioning) large-scale cooperation, as in human societies. Finally, we discuss how individually devised institutions can foster the transition from small-scale to large-scale cooperative groups in human evolution.

Reuse and integration of specification logics: the hybridisation perspective

Hybridisation is a systematic process along which the characteristic features of hybrid logic, both at the syntactic and the semantic levels, are developed on top of an arbitrary logic framed as an institution. It also captures the construction of first-order encodings of such hybridised institutions into theories in first-order logic. The method was originally developed to build suitable logics for the specification of reconfigurable software systems on top of whatever logic is used to describe local requirements of each system’s configuration. Hybridisation has, however, a broader scope, providing a fresh example of yet another development in combining and reusing logics driven by a problem from Computer Science. This paper offers an overview of this method, proposes some new extensions, namely the introduction of full quantification leading to the specification of dynamic modalities, and exemplifies its potential through a didactical application. It is discussed how hybridisation can be successfully used in a formal specification course in which students progress from equational to hybrid specifications in a uniform setting, integrating paradigms, combining data and behaviour, and dealing appropriately with systems evolution and reconfiguration.

A critical analysis of e-government evaluation models at national and local municipal levels

Abstract: The importance of e-government models lies in their offering a basis to measure and guide e-government. There is still no agreement on how to assess a government online. Most of the e-government models are not based on research, nor are they validated. In most countries, e-government has not reached higher stages of growth. Several scholars have shown a confusing picture of e-government. What is lacking is an in-depth analysis of e-government models. Responding to the need for such an analysis, this study identifies the strengths and weaknesses of major national and local e-government evaluation models. The common limitations of most models are focusing on the government and not the citizen, missing qualitative measures, constructing the e-equivalent of a bureaucratic administration, and defining general criteria without sufficient validations. In addition, this study has found that the metrics defined for national e-government are not suitable for municipalities, and most of the existing studies have focused on national e-governments even though local ones are closer to citizens. There is a need for developing a good theoretical model for both national and local municipal e-government.

Problem-Based SRS: método para especificação de requisitos de software baseado em problemas

Requirements specification has long been recognized as critical activity in software development processes because of its impact on project risks when poorly performed. A large amount of studies addresses theoretical aspects, propositions of techniques, and recommended practices for Requirements Engineering (RE). To be successful, RE have to ensure that the specified requirements are complete and correct what means that all intents of the stakeholders in a given business context are covered by the requirements and that no unnecessary requirement was introduced. However, the accurate capture the business intents of the stakeholders remains a challenge and it is a major factor of software project failures. This master’s dissertation presents a novel method referred to as “Problem-Based SRS” aiming at improving the quality of the Software Requirements Specification (SRS) in the sense that the stated requirements provide suitable answers to real customer ́s businesses issues. In this approach, the knowledge about the software requirements is constructed from the knowledge about the customer ́s problems. Problem-Based SRS consists in an organization of activities and outcome objects through a process that contains five main steps. It aims at supporting the software requirements engineering team to systematically analyze the business context and specify the software requirements, taking also into account a first glance and vision of the software. The quality aspects of the specifications are evaluated using traceability techniques and axiomatic design principles. The cases studies conducted and presented in this document point out that the proposed method can contribute significantly to improve the software requirements specification.

Inteligência coletiva e gestão do conhecimento: uma aplicação na rede social Facebook

Social networks rely on concepts such as collaboration, cooperation, replication, flow, speed, interaction, engagement, and aim the continuous sharing and resharing of information in support of the permanent social interaction. Facebook, the largest social network in the world, reached, in May 2016, the mark of 1.09 billion active users daily, draining 161.7 million hours of users’ attention to the website every day. These users share 4.75 billion units of content daily. The research presented in this dissertation aims to investigate the management of knowledge and collective intelligence, from the introduction of mechanisms that aim to enable users to manage and organize current information in the feeds from Facebook groups in which they participate, turning Facebook into a collective knowledge and information management device that goes far beyond mere interaction and communication among people. The adoption of Design Science Research methodology is intended to instill the "genes" of collective intelligence, as presented in the literature, in the computational artifact being developed, so that intelligence can be managed and used to create even more knowledge and intelligence to and by the group. The main theoretical contribution of this dissertation is to discuss knowledge management and collective intelligence in a complementary and integrated manner, showing how efforts to obtain one also contribute to leveraging the other.

A comparative study of navigation meshes

International audience

Measurement of Entrepreneurial Dynamism Capabilities and Performance in Collaborative Networks

Part 16: Performance Measurement Systems

Strategic Context, Organizational Features and Network Performances: A Survey on Collaborative Networked Organizations of Italian SMEs

Part 15: Performance Management Frameworks

O comércio electrónico: realidade algarvia

Dissertação de mest. em Gestão Empresarial, Faculdade de Economia, Univ. do Algarve, 2001