912 resultados para Numerical Algorithms and Problems
Resumo:
L’un des problèmes importants en apprentissage automatique est de déterminer la complexité du modèle à apprendre. Une trop grande complexité mène au surapprentissage, ce qui correspond à trouver des structures qui n’existent pas réellement dans les données, tandis qu’une trop faible complexité mène au sous-apprentissage, c’est-à-dire que l’expressivité du modèle est insuffisante pour capturer l’ensemble des structures présentes dans les données. Pour certains modèles probabilistes, la complexité du modèle se traduit par l’introduction d’une ou plusieurs variables cachées dont le rôle est d’expliquer le processus génératif des données. Il existe diverses approches permettant d’identifier le nombre approprié de variables cachées d’un modèle. Cette thèse s’intéresse aux méthodes Bayésiennes nonparamétriques permettant de déterminer le nombre de variables cachées à utiliser ainsi que leur dimensionnalité. La popularisation des statistiques Bayésiennes nonparamétriques au sein de la communauté de l’apprentissage automatique est assez récente. Leur principal attrait vient du fait qu’elles offrent des modèles hautement flexibles et dont la complexité s’ajuste proportionnellement à la quantité de données disponibles. Au cours des dernières années, la recherche sur les méthodes d’apprentissage Bayésiennes nonparamétriques a porté sur trois aspects principaux : la construction de nouveaux modèles, le développement d’algorithmes d’inférence et les applications. Cette thèse présente nos contributions à ces trois sujets de recherches dans le contexte d’apprentissage de modèles à variables cachées. Dans un premier temps, nous introduisons le Pitman-Yor process mixture of Gaussians, un modèle permettant l’apprentissage de mélanges infinis de Gaussiennes. Nous présentons aussi un algorithme d’inférence permettant de découvrir les composantes cachées du modèle que nous évaluons sur deux applications concrètes de robotique. Nos résultats démontrent que l’approche proposée surpasse en performance et en flexibilité les approches classiques d’apprentissage. Dans un deuxième temps, nous proposons l’extended cascading Indian buffet process, un modèle servant de distribution de probabilité a priori sur l’espace des graphes dirigés acycliques. Dans le contexte de réseaux Bayésien, ce prior permet d’identifier à la fois la présence de variables cachées et la structure du réseau parmi celles-ci. Un algorithme d’inférence Monte Carlo par chaîne de Markov est utilisé pour l’évaluation sur des problèmes d’identification de structures et d’estimation de densités. Dans un dernier temps, nous proposons le Indian chefs process, un modèle plus général que l’extended cascading Indian buffet process servant à l’apprentissage de graphes et d’ordres. L’avantage du nouveau modèle est qu’il admet les connections entres les variables observables et qu’il prend en compte l’ordre des variables. Nous présentons un algorithme d’inférence Monte Carlo par chaîne de Markov avec saut réversible permettant l’apprentissage conjoint de graphes et d’ordres. L’évaluation est faite sur des problèmes d’estimations de densité et de test d’indépendance. Ce modèle est le premier modèle Bayésien nonparamétrique permettant d’apprendre des réseaux Bayésiens disposant d’une structure complètement arbitraire.
Resumo:
In his article ‘Narrative Matters’, Gordon Bates (Bates, 2016) discusses the relevance of the humanities, and in particular that of fiction, in helping us to understand the experiences and problems of adolescents suffering men- tal health problems. Notably, the number of novels pub- lished for children and adolescents that focus on mental health problems has risen considerably since the begin- ning of this century. The ‘Goodreads’ (Goodreads, 2016) website lists more than 1000 fiction titles on matters pertaining to mental illness published for the UK and US markets since 2000.
Resumo:
General-purpose parallel processing for solving day-to-day industrial problems has been slow to develop, partly because of the lack of suitable hardware from well-established, mainstream computer manufacturers and suitably parallelized application software. The parallelization of a CFD-(computational fluid dynamics) flow solution code is known as ESAUNA. This code is part of SAUNA, a large CFD suite aimed at computing the flow around very complex aircraft configurations including complete aircraft. A novel feature of the SAUNA suite is that it is designed to use either block-structured hexahedral grids, unstructured tetrahedral grids, or a hybrid combination of both grid types. ESAUNA is designed to solve the Euler equations or the Navier-Stokes equations, the latter in conjunction with various turbulence models. Two fundamental parallelization concepts are used—namely, grid partitioning and encapsulation of communications. Grid partitioning is applied to both block-structured grid modules and unstructured grid modules. ESAUNA can also be coupled with other simulation codes for multidisciplinary computations such as flow simulations around an aircraft coupled with flutter prediction for transient flight simulations.
Resumo:
This dissertation contains four essays that all share a common purpose: developing new methodologies to exploit the potential of high-frequency data for the measurement, modeling and forecasting of financial assets volatility and correlations. The first two chapters provide useful tools for univariate applications while the last two chapters develop multivariate methodologies. In chapter 1, we introduce a new class of univariate volatility models named FloGARCH models. FloGARCH models provide a parsimonious joint model for low frequency returns and realized measures, and are sufficiently flexible to capture long memory as well as asymmetries related to leverage effects. We analyze the performances of the models in a realistic numerical study and on the basis of a data set composed of 65 equities. Using more than 10 years of high-frequency transactions, we document significant statistical gains related to the FloGARCH models in terms of in-sample fit, out-of-sample fit and forecasting accuracy compared to classical and Realized GARCH models. In chapter 2, using 12 years of high-frequency transactions for 55 U.S. stocks, we argue that combining low-frequency exogenous economic indicators with high-frequency financial data improves the ability of conditionally heteroskedastic models to forecast the volatility of returns, their full multi-step ahead conditional distribution and the multi-period Value-at-Risk. Using a refined version of the Realized LGARCH model allowing for time-varying intercept and implemented with realized kernels, we document that nominal corporate profits and term spreads have strong long-run predictive ability and generate accurate risk measures forecasts over long-horizon. The results are based on several loss functions and tests, including the Model Confidence Set. Chapter 3 is a joint work with David Veredas. We study the class of disentangled realized estimators for the integrated covariance matrix of Brownian semimartingales with finite activity jumps. These estimators separate correlations and volatilities. We analyze different combinations of quantile- and median-based realized volatilities, and four estimators of realized correlations with three synchronization schemes. Their finite sample properties are studied under four data generating processes, in presence, or not, of microstructure noise, and under synchronous and asynchronous trading. The main finding is that the pre-averaged version of disentangled estimators based on Gaussian ranks (for the correlations) and median deviations (for the volatilities) provide a precise, computationally efficient, and easy alternative to measure integrated covariances on the basis of noisy and asynchronous prices. Along these lines, a minimum variance portfolio application shows the superiority of this disentangled realized estimator in terms of numerous performance metrics. Chapter 4 is co-authored with Niels S. Hansen, Asger Lunde and Kasper V. Olesen, all affiliated with CREATES at Aarhus University. We propose to use the Realized Beta GARCH model to exploit the potential of high-frequency data in commodity markets. The model produces high quality forecasts of pairwise correlations between commodities which can be used to construct a composite covariance matrix. We evaluate the quality of this matrix in a portfolio context and compare it to models used in the industry. We demonstrate significant economic gains in a realistic setting including short selling constraints and transaction costs.
Resumo:
Americans are accustomed to a wide range of data collection in their lives: census, polls, surveys, user registrations, and disclosure forms. When logging onto the Internet, users’ actions are being tracked everywhere: clicking, typing, tapping, swiping, searching, and placing orders. All of this data is stored to create data-driven profiles of each user. Social network sites, furthermore, set the voluntarily sharing of personal data as the default mode of engagement. But people’s time and energy devoted to creating this massive amount of data, on paper and online, are taken for granted. Few people would consider their time and energy spent on data production as labor. Even if some people do acknowledge their labor for data, they believe it is accessory to the activities at hand. In the face of pervasive data collection and the rising time spent on screens, why do people keep ignoring their labor for data? How has labor for data been become invisible, as something that is disregarded by many users? What does invisible labor for data imply for everyday cultural practices in the United States? Invisible Labor for Data addresses these questions. I argue that three intertwined forces contribute to framing data production as being void of labor: data production institutions throughout history, the Internet’s technological infrastructure (especially with the implementation of algorithms), and the multiplication of virtual spaces. There is a common tendency in the framework of human interactions with computers to deprive data and bodies of their materiality. My Introduction and Chapter 1 offer theoretical interventions by reinstating embodied materiality and redefining labor for data as an ongoing process. The middle Chapters present case studies explaining how labor for data is pushed to the margin of the narratives about data production. I focus on a nationwide debate in the 1960s on whether the U.S. should build a databank, contemporary Big Data practices in the data broker and the Internet industries, and the group of people who are hired to produce data for other people’s avatars in the virtual games. I conclude with a discussion on how the new development of crowdsourcing projects may usher in the new chapter in exploiting invisible and discounted labor for data.
Resumo:
Understanding how biodiversity spatially distribute over both the short term and long term, and what factors are affecting the distribution, are critical for modeling the spatial pattern of biodiversity as well as for promoting effective conservation planning and practices. This dissertation aims to examine factors that influence short-term and long-term avian distribution from the geographical sciences perspective. The research develops landscape level habitat metrics to characterize forest height heterogeneity and examines their efficacies in modelling avian richness at the continental scale. Two types of novel vegetation-height-structured habitat metrics are created based on second order texture algorithms and the concepts of patch-based habitat metrics. I correlate the height-structured metrics with the richness of different forest guilds, and also examine their efficacies in multivariate richness models. The results suggest that height heterogeneity, beyond canopy height alone, supplements habitat characterization and richness models of two forest bird guilds. The metrics and models derived in this study demonstrate practical examples of utilizing three-dimensional vegetation data for improved characterization of spatial patterns in species richness. The second and the third projects focus on analyzing centroids of avian distributions, and testing hypotheses regarding the direction and speed of these shifts. I first showcase the usefulness of centroids analysis for characterizing the distribution changes of a few case study species. Applying the centroid method on 57 permanent resident bird species, I show that multi-directional distribution shifts occurred in large number of studied species. I also demonstrate, plain birds are not shifting their distribution faster than mountain birds, contrary to the prediction based on climate change velocity hypothesis. By modelling the abundance change rate at regional level, I show that extreme climate events and precipitation measures associate closely with some of the long-term distribution shifts. This dissertation improves our understanding on bird habitat characterization for species richness modelling, and expands our knowledge on how avian populations shifted their ranges in North America responding to changing environments in the past four decades. The results provide an important scientific foundation for more accurate predictive species distribution modeling in future.
Resumo:
The present chapter discusses the assets model as a theoretical approach to the study of health behavior and health promotion. The model emphasizes people’s talents, competences, and resources. In this chapter, a health asset is defined as any factor or resource that maximizes the opportunities for individuals, local communities, and populations to attain and maintain health and well-being. This perspective expands and complements the current medical model as it focuses on the development of a sense of empowerment in community members to prevent and manage their own health. Therefore, in this chapter we address the concepts of salutogenesis, social support, resilience, coping, self-regulation, social capital, and personal and social competence, which are central to the development of individuals’ potential to manage and savor their own health, creating the conditions for self-fulfillment. Additionally, we demonstrate how the assets model guides the study of children’s and adolescents’ health in the Portuguese Health Behaviour in School-aged Children study (www.hbsc.org), concentrating on areas such as active lifestyles and quality-of-life perception. Finally, we present a roadmap for action that emphasizes the need to identify the factors that make children and adolescents happy and healthy individuals, while minimizing risks and problems they naturally encounter throughout their development. We also argue for the need to involve young people in discussions concerning their health and health promotion practices, focusing on the development of talents, capabilities, and positive expectations for the future.
Resumo:
HERMES is one of the projects in the European ATT Programme. The ATT Programme (or DRIVE II as it is frequently referred to) is an application oriented Community Research and Technological Development Programme that has been conceived and implemented with the objective of contributing to the competitiveness of Europe and to its social and economic cohesion. An important means toward this end is the direct collaboration between different European sector actors: road authorities, fleet operators, road user representatives, industry, and research institutions. DRIVE I has already achieved an important step into this direction. DRIVE II aims at providing a framework that encourages even closer cooperation through large scale international pilot projects that will require common functional and technical specifications for the systems to be implemented at least between the partners directly involved in any project. HERMES is one of the so-called "supporting R&D projects" that provides strategies, algorithms and systems for the pilot applications
Resumo:
Resource allocation decisions are made to serve the current emergency without knowing which future emergency will be occurring. Different ordered combinations of emergencies result in different performance outcomes. Even though future decisions can be anticipated with scenarios, previous models follow an assumption that events over a time interval are independent. This dissertation follows an assumption that events are interdependent, because speed reduction and rubbernecking due to an initial incident provoke secondary incidents. The misconception that secondary incidents are not common has resulted in overlooking a look-ahead concept. This dissertation is a pioneer in relaxing the structural assumptions of independency during the assignment of emergency vehicles. When an emergency is detected and a request arrives, an appropriate emergency vehicle is immediately dispatched. We provide tools for quantifying impacts based on fundamentals of incident occurrences through identification, prediction, and interpretation of secondary incidents. A proposed online dispatching model minimizes the cost of moving the next emergency unit, while making the response as close to optimal as possible. Using the look-ahead concept, the online model flexibly re-computes the solution, basing future decisions on present requests. We introduce various online dispatching strategies with visualization of the algorithms, and provide insights on their differences in behavior and solution quality. The experimental evidence indicates that the algorithm works well in practice. After having served a designated request, the available and/or remaining vehicles are relocated to a new base for the next emergency. System costs will be excessive if delay regarding dispatching decisions is ignored when relocating response units. This dissertation presents an integrated method with a principle of beginning with a location phase to manage initial incidents and progressing through a dispatching phase to manage the stochastic occurrence of next incidents. Previous studies used the frequency of independent incidents and ignored scenarios in which two incidents occurred within proximal regions and intervals. The proposed analytical model relaxes the structural assumptions of Poisson process (independent increments) and incorporates evolution of primary and secondary incident probabilities over time. The mathematical model overcomes several limiting assumptions of the previous models, such as no waiting-time, returning rule to original depot, and fixed depot. The temporal locations flexible with look-ahead are compared with current practice that locates units in depots based on Poisson theory. A linearization of the formulation is presented and an efficient heuristic algorithm is implemented to deal with a large-scale problem in real-time.
Resumo:
Over the last few years, more and more heuristic decision making techniques have been inspired by nature, e.g. evolutionary algorithms, ant colony optimisation and simulated annealing. More recently, a novel computational intelligence technique inspired by immunology has emerged, called Artificial Immune Systems (AIS). This immune system inspired technique has already been useful in solving some computational problems. In this keynote, we will very briefly describe the immune system metaphors that are relevant to AIS. We will then give some illustrative real-world problems suitable for AIS use and show a step-by-step algorithm walkthrough. A comparison of AIS to other well-known algorithms and areas for future work will round this keynote off. It should be noted that as AIS is still a young and evolving field, there is not yet a fixed algorithm template and hence actual implementations might differ somewhat from the examples given here
Resumo:
International audience
Resumo:
Reliability and dependability modeling can be employed during many stages of analysis of a computing system to gain insights into its critical behaviors. To provide useful results, realistic models of systems are often necessarily large and complex. Numerical analysis of these models presents a formidable challenge because the sizes of their state-space descriptions grow exponentially in proportion to the sizes of the models. On the other hand, simulation of the models requires analysis of many trajectories in order to compute statistically correct solutions. This dissertation presents a novel framework for performing both numerical analysis and simulation. The new numerical approach computes bounds on the solutions of transient measures in large continuous-time Markov chains (CTMCs). It extends existing path-based and uniformization-based methods by identifying sets of paths that are equivalent with respect to a reward measure and related to one another via a simple structural relationship. This relationship makes it possible for the approach to explore multiple paths at the same time,· thus significantly increasing the number of paths that can be explored in a given amount of time. Furthermore, the use of a structured representation for the state space and the direct computation of the desired reward measure (without ever storing the solution vector) allow it to analyze very large models using a very small amount of storage. Often, path-based techniques must compute many paths to obtain tight bounds. In addition to presenting the basic path-based approach, we also present algorithms for computing more paths and tighter bounds quickly. One resulting approach is based on the concept of path composition whereby precomputed subpaths are composed to compute the whole paths efficiently. Another approach is based on selecting important paths (among a set of many paths) for evaluation. Many path-based techniques suffer from having to evaluate many (unimportant) paths. Evaluating the important ones helps to compute tight bounds efficiently and quickly.
Resumo:
Over the last decade, rapid development of additive manufacturing techniques has allowed the fabrication of innovative and complex designs. One field that can benefit from such technology is heat exchanger fabrication, as heat exchanger design has become more and more complex due to the demand for higher performance particularly on the air side of the heat exchanger. By employing the additive manufacturing, a heat exchanger design was successfully realized, which otherwise would have been very difficult to fabricate using conventional fabrication technologies. In this dissertation, additive manufacturing technique was implemented to fabricate an advanced design which focused on a combination of heat transfer surface and fluid distribution system. Although the application selected in this dissertation is focused on power plant dry cooling applications, the results of this study can directly and indirectly benefit other sectors as well, as the air-side is often the limiting side for in liquid or single phase cooling applications. Two heat exchanger designs were studied. One was an advanced metallic heat exchanger based on manifold-microchannel technology and the other was a polymer heat exchanger based on utilization of prime surface technology. Polymer heat exchangers offer several advantages over metals such as antifouling, anticorrosion, lightweight and often less expensive than comparable metallic heat exchangers. A numerical modeling and optimization were performed to calculate a design that yield an optimum performance. The optimization results show that significant performance enhancement is noted compared to the conventional heat exchangers like wavy fins and plain plate fins. Thereafter, both heat exchangers were scaled down and fabricated using additive manufacturing and experimentally tested. The manifold-micro channel design demonstrated that despite some fabrication inaccuracies, compared to a conventional wavy-fin surface, 15% - 50% increase in heat transfer coefficient was possible for the same pressure drop value. In addition, if the fabrication inaccuracy can be eliminated, an even larger performance enhancement is predicted. Since metal based additive manufacturing is still in the developmental stage, it is anticipated that with further refinement of the manufacturing process in future designs, the fabrication accuracy can be improved. For the polymer heat exchanger, by fabricating a very thin wall heat exchanger (150μm), the wall thermal resistance, which usually becomes the limiting side for polymer heat exchanger, was calculated to account for only up to 3% of the total thermal resistance. A comparison of air-side heat transfer coefficient of the polymer heat exchanger with some of the commercially available plain plate fin surface heat exchangers show that polymer heat exchanger performance is equal or superior to plain plate fin surfaces. This shows the promising potential for polymer heat exchangers to compete with conventional metallic heat exchangers when an additive manufacturing-enabled fabrication is utilized. Major contributions of this study are as follows: (1) For the first time demonstrated the potential of additive manufacturing in metal printing of heat exchangers that benefit from a sophisticated design to yield a performance substantially above the respective conventional systems. Such heat exchangers cannot be fabricated with the conventional fabrication techniques. (2) For the first time demonstrated the potential of additive manufacturing to produce polymer heat exchangers that by design minimize the role of thermal conductivity and deliver a thermal performance equal or better that their respective metallic heat exchangers. In addition of other advantages of polymer over metal like antifouling, anticorrosion, and lightweight. Details of the work are documented in respective chapters of this thesis.
Resumo:
libtissue is a software system for implementing and testing AIS algorithms on real-world computer security problems. AIS algorithms are implemented as a collection of cells, antigen and signals interacting within a tissue compartment. Input data to the tissue comes in the form of realtime events generated by sensors monitoring a system under surveillance, and cells are actively able to affect the monitored system through response mechanisms. libtissue is being used by researchers on a project at the University of Nottingham to explore the application of a range of immune-inspired algorithms to problems in intrusion detection. This talk describes the architecture and design of libtissue, along with the implementation of a simple algorithm and its application to a computer security problem.
Resumo:
Over the last few years, more and more heuristic decision making techniques have been inspired by nature, e.g. evolutionary algorithms, ant colony optimisation and simulated annealing. More recently, a novel computational intelligence technique inspired by immunology has emerged, called Artificial Immune Systems (AIS). This immune system inspired technique has already been useful in solving some computational problems. In this keynote, we will very briefly describe the immune system metaphors that are relevant to AIS. We will then give some illustrative real-world problems suitable for AIS use and show a step-by-step algorithm walkthrough. A comparison of AIS to other well-known algorithms and areas for future work will round this keynote off. It should be noted that as AIS is still a young and evolving field, there is not yet a fixed algorithm template and hence actual implementations might differ somewhat from the examples given here.