[pt] O forward premium puzzle é um dos puzzles mais notáveis no mercado cambial. Seus estudos tiveram início na primeira metade da década de 1980 e desde então diversas metodologias, ao longo dos anos, foram elaboradas e utilizadas para tentar explicar a sua ocorrência. Entretanto, apesar dos esforços dos pesquisadores, ainda não há uma solução inequívoca. O objetivo desta dissertação é analisar esta anomalia no contexto macroeconômico dos países emergentes, considerando-se a abordagem das expectativas não-racionais dos agentes. Para isso, foram aplicadas técnicas de aprendizado econométrico, conforme as metodologias propostas por Chakraborty e Evans (2008) e Reed (2019). Segundo estas, as expectativas dos agentes acerca da taxa de câmbio spot futura, são modeladas de acordo com algoritmos de aprendizado e se mantêm próximas à solução de expectativas racionais, porém com desvios gerados por erros de previsão passados. Os resultados alcançados corroboram aqueles encontrados na literatura relacionada, demonstrando que o aprendizado econométrico pode fornecer uma explicação para o forward premium puzzle também quando se considera a taxa cambial das moedas dos países emergentes frente ao Dólar americano, tanto para o modelo com um único estado quanto para o modelo com dois estados. As simulações realizadas reproduzem as principais características empíricas encontradas na amostra analisada e exprimem a importância da persistência dos fundamentos monetários para explicar o viés negativo no coeficiente do forward premium. / [en] The forward premium puzzle is one of the most notable puzzles in the foreign exchange market. Seminal studies in the field began in the first half of the 1980s. Since then, several methodologies have been proposed and tested, aiming to explain the occurrence of forward premium puzzle. However, despite the researchers efforts, no unequivocal solution has been fund. The objective of this dissertation is to analyze this anomaly in the macroeconomic context of emerging countries considering the approach of the non-rational expectations. To that end, adaptive learning techniques were applied, following the methodologies proposed by Chakraborty and Evans (2008) and Reed (2019), where the modeling of agents expectations about the future spot exchange rate is conducted using learning algorithms and remain close to the rational expectations solution, but with deviations generated from past forecast errors. The results corroborate those found in the current body of literature, suggesting that adaptive learning can provide an explanation for the forward premium puzzle also when considering the exchange rate of emerging market currencies against the US dollar, both for the single-state model and for the two-state model. The simulations results reproduce the main empirical features found in the analyzed sample and express the importance of the monetary fundamentals persistence to explain the negative bias in the forward premium coefficient.

User-centered computer based learning is an emerging field of interdisciplinary research. Research in diverse areas such as psychology, computer science, neuroscience and signal processing is making contributions the promise to take this field to the next level. Learning systems built using contributions from these fields could be used in actual training and education instead of just laboratory proof-of-concept. One of the important advances in this research is the detection and assessment of the cognitive and emotional state of the learner using such systems. This capability moves development beyond the use of traditional user performance metrics to include system intelligence measures that are based on current neuroscience theories. These advances are of paramount importance in the success and wide spread use of learning systems that are automated and intelligent. Emotion is considered an important aspect of how learning occurs, and yet estimating it and making adaptive adjustments are not part of most learning systems. In this research we focus on one specific aspect of constructing an adaptive and intelligent learning system, that is, estimation of the emotion of the learner as he/she is using the automated training system. The challenge starts with the definition of the emotion and the utility of it in human life. The next challenge is to measure the co-varying factors of the emotions in a non-invasive way, and find consistent features from these measures that are valid across wide population. In this research we use four physiological sensors that are non-invasive, and establish a methodology of utilizing the data from these sensors using different signal processing tools. A validated set of visual stimuli used worldwide in the research of emotion and attention, called International Affective Picture System (IAPS), is used. A dataset is collected from the sensors in an experiment designed to elicit emotions from these validated visual stimuli. We describe a novel wavelet method to calculate hemispheric asymmetry metric using electroencephalography data. This method is tested against typically used power spectral density method. We show overall improvement in accuracy in classifying specific emotions using the novel method. We also show distinctions between different discrete emotions from the autonomic nervous system activity using electrocardiography, electrodermal activity and pupil diameter changes. Findings from different features from these sensors are used to give guidelines to use each of the individual sensors in the adaptive learning environment.

Sistemas de Aprendizagem Adaptativos e Inteligentes, tema de pesquisa recente no mundo, são ambientes com arquitetura e algoritmos específicos, que consideram as características individuais de cada estudante para selecionar o objeto de aprendizagem mais adequado a ser oferecido ao aluno. O rápido desenvolvimento da infraestrutura sem fio e o amplo uso de dispositivos móveis na vida diária das pessoas motivam as pesquisas relativas ao uso desses dispositivos na educação, proporcionando o m-learning. Assim, relacionado a essas linhas de pesquisa, este trabalho propõe a arquitetura AdaptMLearning, elaborada para prover a aprendizagem em plataformas móveis e não móveis, considerando a seleção de objetos de aprendizagem que melhor se adaptam a diversos aspectos, tais como: dados sobre a tecnologia utilizada para acesso; informações sobre o estilo de aprendizagem de um estudante; desempenho e tempo associados à interação do estudante com o objeto de aprendizagem; conhecimentos adquiridos pelo estudante em consonância ao conteúdo do curso; e a garantia de que não só o professor possa configurar as adaptações a serem oferecidas ao seu curso, como também o aluno tenha a possibilidade de informar sua preferência pelos tipos de mídia. Essa arquitetura é baseada no modelo de referência AHAM para sistemas adaptativos AEHS, contemplando a quádrupla: espaço do conhecimento, modelo do usuário, observações e modelo de adaptação, referente à definição lógica desses sistemas. Na AdaptMLearning, foram desenvolvidos alguns algoritmos, utilizando-se o modelo FSLSM, relacionado aos estilos de aprendizagem de um estudante e o padrão IEEE 1484 para catalogação dos objetos de aprendizagem e uso de alguns atributos de suas categorias, associados às dimensões dos estilos de aprendizagem do modelo FSLSM. O algoritmo calcula um peso para um objeto catalogado em cada dimensão e permite uma busca pelo objeto mais adequado ao estilo do estudante, além de usar a computação fuzzy, para avaliar se o estudante pode sofrer mudanças no seu estilo, deve receber reforço ou necessita de um reestudo em determinado assunto de um curso, por meio de resultados obtidos com o tempo de estudo e desempenho. Também, este trabalho apresenta o desenvolvimento e a avaliação de um simulador para a arquitetura AdaptMLearning e seus algoritmos, realizada utilizando diversos cenários de simulação, envolvendo estudantes, cursos e tecnologias com diferentes configurações. Assim sendo, com base nos resultados obtidos por meio da avaliação, foi possível discutir, analisar e identificar o potencial de uso da AdaptMLearning e de seus algoritmos em uma situação real para elaboração de um ambiente de aprendizagem ou agregação a um ambiente existente. / Intelligent and Adaptive Learning Systems, subject of recent research in the world, are environments with specific architectures and algorithms, designed considering the individual characteristics of each student. The rapid development of wireless infrastructures and wide use of mobile devices in people\'s everyday life encourage research about the use of these devices in education, providing the mlearning. In the context of such research, this work proposes the AdaptMLearning architecture that was designed to be a learning infrastructure for mobile and nonmobile platforms. This architecture provides a selection of learning objects that takes into account as adaptation criteria the following data: the mobile device\'s technological specification; the student\'s learning style information, his/her performance and spent time associated to the student\'s interaction with the learning object; previously acquired knowledge by the student related to the course\'s content. In addition, it also allows the teacher to interfere in the adaptation criteria used during the study simulation, and allows the student to indicate his/her preferences for media types. This architecture is based on AHAM reference model for adaptive systems AEHS and uses the quadruple: the knowledge space, the user model, the observations and the model adaptation, referring to the logical definition of these systems. To implement the AdaptMLearning architecture some algorithms using the FSLSM model related to the student\'s learning styles were developed. The algorithms use the IEEE 1484 for cataloging learning objects and some of its categories and attributes associated with dimensions of learning styles FSLSM model, are used to compute a weight of an object in each dimension allowing a search of the most appropriate object according to the student\'s learning styles; and the use of fuzzy computing, considering that the student\'s learning style can change, determines if the student has to receive reinforcement or need a new study in a particular subject of a course, when the student gets unsatisfactory results in terms of timing and performance in a course\'s subject. Also, this work also presents the development and evaluation of a simulator for the AdaptMLearning architecture and their algorithms. The evaluation of the simulator was done by means of many simulations scenarios, considering students, courses and technologies with different settings. Based on the results obtained from the evaluation it was possible to discuss, analyze and identify the potential use of AdaptMLearning architecture and their algorithms in a real situation for developing a learning environment or its aggregation to an existing environment.

Sistemas de Aprendizagem Adaptativos e Inteligentes, tema de pesquisa recente no mundo, são ambientes com arquitetura e algoritmos específicos, que consideram as características individuais de cada estudante para selecionar o objeto de aprendizagem mais adequado a ser oferecido ao aluno. O rápido desenvolvimento da infraestrutura sem fio e o amplo uso de dispositivos móveis na vida diária das pessoas motivam as pesquisas relativas ao uso desses dispositivos na educação, proporcionando o m-learning. Assim, relacionado a essas linhas de pesquisa, este trabalho propõe a arquitetura AdaptMLearning, elaborada para prover a aprendizagem em plataformas móveis e não móveis, considerando a seleção de objetos de aprendizagem que melhor se adaptam a diversos aspectos, tais como: dados sobre a tecnologia utilizada para acesso; informações sobre o estilo de aprendizagem de um estudante; desempenho e tempo associados à interação do estudante com o objeto de aprendizagem; conhecimentos adquiridos pelo estudante em consonância ao conteúdo do curso; e a garantia de que não só o professor possa configurar as adaptações a serem oferecidas ao seu curso, como também o aluno tenha a possibilidade de informar sua preferência pelos tipos de mídia. Essa arquitetura é baseada no modelo de referência AHAM para sistemas adaptativos AEHS, contemplando a quádrupla: espaço do conhecimento, modelo do usuário, observações e modelo de adaptação, referente à definição lógica desses sistemas. Na AdaptMLearning, foram desenvolvidos alguns algoritmos, utilizando-se o modelo FSLSM, relacionado aos estilos de aprendizagem de um estudante e o padrão IEEE 1484 para catalogação dos objetos de aprendizagem e uso de alguns atributos de suas categorias, associados às dimensões dos estilos de aprendizagem do modelo FSLSM. O algoritmo calcula um peso para um objeto catalogado em cada dimensão e permite uma busca pelo objeto mais adequado ao estilo do estudante, além de usar a computação fuzzy, para avaliar se o estudante pode sofrer mudanças no seu estilo, deve receber reforço ou necessita de um reestudo em determinado assunto de um curso, por meio de resultados obtidos com o tempo de estudo e desempenho. Também, este trabalho apresenta o desenvolvimento e a avaliação de um simulador para a arquitetura AdaptMLearning e seus algoritmos, realizada utilizando diversos cenários de simulação, envolvendo estudantes, cursos e tecnologias com diferentes configurações. Assim sendo, com base nos resultados obtidos por meio da avaliação, foi possível discutir, analisar e identificar o potencial de uso da AdaptMLearning e de seus algoritmos em uma situação real para elaboração de um ambiente de aprendizagem ou agregação a um ambiente existente. / Intelligent and Adaptive Learning Systems, subject of recent research in the world, are environments with specific architectures and algorithms, designed considering the individual characteristics of each student. The rapid development of wireless infrastructures and wide use of mobile devices in people\'s everyday life encourage research about the use of these devices in education, providing the mlearning. In the context of such research, this work proposes the AdaptMLearning architecture that was designed to be a learning infrastructure for mobile and nonmobile platforms. This architecture provides a selection of learning objects that takes into account as adaptation criteria the following data: the mobile device\'s technological specification; the student\'s learning style information, his/her performance and spent time associated to the student\'s interaction with the learning object; previously acquired knowledge by the student related to the course\'s content. In addition, it also allows the teacher to interfere in the adaptation criteria used during the study simulation, and allows the student to indicate his/her preferences for media types. This architecture is based on AHAM reference model for adaptive systems AEHS and uses the quadruple: the knowledge space, the user model, the observations and the model adaptation, referring to the logical definition of these systems. To implement the AdaptMLearning architecture some algorithms using the FSLSM model related to the student\'s learning styles were developed. The algorithms use the IEEE 1484 for cataloging learning objects and some of its categories and attributes associated with dimensions of learning styles FSLSM model, are used to compute a weight of an object in each dimension allowing a search of the most appropriate object according to the student\'s learning styles; and the use of fuzzy computing, considering that the student\'s learning style can change, determines if the student has to receive reinforcement or need a new study in a particular subject of a course, when the student gets unsatisfactory results in terms of timing and performance in a course\'s subject. Also, this work also presents the development and evaluation of a simulator for the AdaptMLearning architecture and their algorithms. The evaluation of the simulator was done by means of many simulations scenarios, considering students, courses and technologies with different settings. Based on the results obtained from the evaluation it was possible to discuss, analyze and identify the potential use of AdaptMLearning architecture and their algorithms in a real situation for developing a learning environment or its aggregation to an existing environment.

Leadership theories have evolved since their inception and leadership development should also evolve in response to the changing needs of their organizations. There is a gap in the literature on the functionality of the various 360-feedback platforms that are mentioned in literature which makes it difficult to understand what reporting functionality is available in existing systems. The goal of this study was to examine participants' views of a newly developed Adaptive Learning 360-feedback (AL360) treatment platform for leadership development with a focus on self-directed learning for improved accountability. Participants were asked to share their learning development experience in the areas of person, process, and context with suggestions for improvement and how did this development compare to any other leadership development that they had experienced in the past. The AL360 platform was built by applying the research findings of scholars in the areas of leadership, 360-feedback leadership development and educational technologies. The treatment was applied in a business setting for all levels of leadership development. The case study identified creation of leadership of self-awareness and accountability for development without significant intervention of an executive coach. The research findings also list improvement opportunities, limitations, and future considerations.

In this thesis, research efforts are dedicated towards the development of practical adaptable techniques to be used in Smart Homes and Buildings, with the aim to improve energy management and conservation, while enhancing the learning capabilities of Programmable Communicating Thermostats (PCT) – “transforming” them into smart adaptable devices, i.e., “Smart Thermostats”. An Adaptable Hybrid Intelligent System utilizing Wireless Sensor Network (WSN) and Artificial Intelligence (AI) techniques is presented, based on which, a novel Adaptive Learning System (ALS) model utilizing WSN, a rule-based system and Adaptive Resonance Theory (ART) concepts is proposed. The main goal of the ALS is to adapt to the occupant’s pattern and/or schedule changes by providing comfort, while not ignoring the energy conservation aspect. The proposed ALS analytical model is a technique which enables PCTs to learn and adapt to user input pattern changes and/or other parameters of interest. A new algorithm for finding the global maximum in a predefined interval within a two dimensional space is proposed. The proposed algorithm is a synergy of reward/punish concepts from the reinforcement learning (RL) and agent-based technique, for use in small-scale embedded systems with limited memory and/or processing power, such as the wireless sensor/actuator nodes. An application is implemented to observe the algorithm at work and to demonstrate its main features. It was observed that the “RL and Agent-based Search”, versus the “RL only” technique, yielded better performance results with respect to the number of iterations and function evaluations needed to find the global maximum. Furthermore, a “House Simulator” is developed as a tool to simulate house heating/cooling systems and to assist in the practical implementation of the ALS model under different scenarios. The main building blocks of the simulator are the “House Simulator”, the “Smart Thermostat”, and a placeholder for the “Adaptive Learning Models”. As a result, a novel adaptive learning algorithm, “Observe, Learn and Adapt” (OLA) is proposed and demonstrated, reflecting the main features of the ALS model. Its evaluation is achieved with the aid of the “House Simulator”. OLA, with the use of sensors and the application of the ALS model learning technique, captures the essence of an actual PCT reflecting a smart and adaptable device. The experimental performance results indicate adaptability and potential energy savings of the single in comparison to the zone controlled scenarios with the OLA capabilities being enabled.

In this thesis, research efforts are dedicated towards the development of practical adaptable techniques to be used in Smart Homes and Buildings, with the aim to improve energy management and conservation, while enhancing the learning capabilities of Programmable Communicating Thermostats (PCT) – “transforming” them into smart adaptable devices, i.e., “Smart Thermostats”. An Adaptable Hybrid Intelligent System utilizing Wireless Sensor Network (WSN) and Artificial Intelligence (AI) techniques is presented, based on which, a novel Adaptive Learning System (ALS) model utilizing WSN, a rule-based system and Adaptive Resonance Theory (ART) concepts is proposed. The main goal of the ALS is to adapt to the occupant’s pattern and/or schedule changes by providing comfort, while not ignoring the energy conservation aspect. The proposed ALS analytical model is a technique which enables PCTs to learn and adapt to user input pattern changes and/or other parameters of interest. A new algorithm for finding the global maximum in a predefined interval within a two dimensional space is proposed. The proposed algorithm is a synergy of reward/punish concepts from the reinforcement learning (RL) and agent-based technique, for use in small-scale embedded systems with limited memory and/or processing power, such as the wireless sensor/actuator nodes. An application is implemented to observe the algorithm at work and to demonstrate its main features. It was observed that the “RL and Agent-based Search”, versus the “RL only” technique, yielded better performance results with respect to the number of iterations and function evaluations needed to find the global maximum. Furthermore, a “House Simulator” is developed as a tool to simulate house heating/cooling systems and to assist in the practical implementation of the ALS model under different scenarios. The main building blocks of the simulator are the “House Simulator”, the “Smart Thermostat”, and a placeholder for the “Adaptive Learning Models”. As a result, a novel adaptive learning algorithm, “Observe, Learn and Adapt” (OLA) is proposed and demonstrated, reflecting the main features of the ALS model. Its evaluation is achieved with the aid of the “House Simulator”. OLA, with the use of sensors and the application of the ALS model learning technique, captures the essence of an actual PCT reflecting a smart and adaptable device. The experimental performance results indicate adaptability and potential energy savings of the single in comparison to the zone controlled scenarios with the OLA capabilities being enabled.

This study aimed to examine the effectiveness of video self-modelling (VSM) and the iPad application (Fast Fact Math, FFM) interventions on a group of Year 5 students to increase their knowledge of basic number facts. This study also aimed to measure the intervention group’s self-efficacy levels (Patterns of Adaptive Learning Scales, PALS) before and following the interventions. Participants were drawn from a decile 9 primary school in a suburban area (teaches Year 1 to Year 6). The Test (pre-, mid-, and post-test phases) were administered by a class teacher to all Year 5 students. Following consultation with the teacher, eight students whose scores fell below the 25th percentile were invited to participate in the study. The intervention group took a specific level test to ascertain their basic number facts performance on all four operations (addition, subtraction, multiplication, and division). The videos and the FFM app were personalised to each intervention group’s members in an effort to elicit from the errors that they made on specific level test. At the completion of each intervention sessions, session probes were conducted. Meanwhile, the researcher gave a self-efficacy test (PALS) to the participants before and following intervention phases. Results showed that, although more than half of the intervention group increased their basic number fact performance level following the interventions, their overall self-efficacy rating on PALS did not change. Results also showed that VSM is a time-efficient and rapid learning method to use with the intervention group as opposed to the iPad app, which took two times longer to complete a session. Further areas of study are suggested.

We develop graph-based methods for semi-supervised learning based on label propagation on a data similarity graph. When data is abundant or arrive in a stream, the problems of computation and data storage arise for any graph-based method. We propose a fast approximate online algorithm that solves for the harmonic solution on an approximate graph. We show, both empirically and theoretically, that good behavior can be achieved by collapsing nearby points into a set of local representative points that minimize distortion. Moreover, we regularize the harmonic solution to achieve better stability properties. We also present graph-based methods for detecting conditional anomalies and apply them to the identification of unusual clinical actions in hospitals. Our hypothesis is that patient-management actions that are unusual with respect to the past patients may be due to errors and that it is worthwhile to raise an alert if such a condition is encountered. Conditional anomaly detection extends standard unconditional anomaly framework but also faces new problems known as fringe and isolated points. We devise novel nonparametric graph-based methods to tackle these problems. Our methods rely on graph connectivity analysis and soft harmonic solution. Finally, we conduct an extensive human evaluation study of our conditional anomaly methods by 15 experts in critical care.

This thesis develops flexible and principled nonparametric learning algorithms to explore, understand, and predict high dimensional and complex datasets. Such data appear frequently in modern scientific domains and lead to numerous important applications. For example, exploring high dimensional functional magnetic resonance imaging data helps us to better understand brain functionalities; inferring large-scale gene regulatory network is crucial for new drug design and development; detecting anomalies in high dimensional transaction databases is vital for corporate and government security. Our main results include a rigorous theoretical framework and efficient nonparametric learning algorithms that exploit hidden structures to overcome the curse of dimensionality when analyzing massive high dimensional datasets. These algorithms have strong theoretical guarantees and provide high dimensional nonparametric recipes for many important learning tasks, ranging from unsupervised exploratory data analysis to supervised predictive modeling. In this thesis, we address three aspects: 1 Understanding the statistical theories of high dimensional nonparametric inference, including risk, estimation, and model selection consistency; 2 Designing new methods for different data-analysis tasks, including regression, classification, density estimation, graphical model learning, multi-task learning, spatial-temporal adaptive learning; 3 Demonstrating the usefulness of these methods in scientific applications, including functional genomics, cognitive neuroscience, and meteorology. In the last part of this thesis, we also present the future vision of high dimensional and large-scale nonparametric inference.