Pas de résumé / CMOS image sensor based on Active pixel sensor has considerably contributed to the imaging market and research interest in the past decade. Furthermore technology advancement has provided the capability to integrate more and more functionality into a single chip in multiple layers leading to a new paradigm, 3D integration. CMOS image sensor is one such application which could utilize the capability of 3D stacked architecture to achieve dedicated technologies in different layers, wire length reduction, less area, improved performancesThis research work is focused mainly on the early stages of design space exploration using hierarchical approach and aims at reducing time to market. This work investigates the imager from the top-down design perspective. Methodical anal y sis of imager is performed to achieve high level of flexibility and modularity. Re-useable models are developed to explore early design choices throughout the hierarchy. Finally, pareto front (providing trade off solutions) methodology is applied to explore the operating range of individual block at system level to help the designer making his design choice. Furthermore the thermal issues which get aggravated in the 3D stacked chip on the performance of the imager are studied. Systeme based thermal model is built to investigate the behavior of imager pixel matrix and to simulate the pixel matrix at high speed with acceptable accuracy compared to electrical simulations. The modular nature of the model makes simulations with future matrix extension straightforward. Validation of the thermal model with respect to electrical simulations is discussed. Finally an integrated design flow is developed to perform 3D floorplanning and to perform thermal anal y sis of the imager pixel matrix.

Des problèmes de Bandit constituent une séquence d’allocation dynamique. D’une part, l’agent de système doit explorer son environnement ( à savoir des bras de machine) pour recueillir des informations; d’autre part, il doit exploiter les informations collectées pour augmenter la récompense. Comment d’équilibrer adéquatement la phase d’exploration et la phase d’exploitation, c’est une obscurité des problèmes de Bandit, et la plupart des chercheurs se concentrent des efforts sur les stratégies d’équilibration entre l’exploration et l’exploitation. Dans cette dissertation, nous nous concentrons sur l’étude de deux problèmes spécifiques de Bandit: les problèmes de Bandit contextuel et les problèmes de Bandit Multi- objectives. Cette dissertation propose deux aspects de contributions. La première concerne la classification sous la surveillance partielle, laquelle nous codons comme le problème de Bandit contextuel avec des informations partielles. Ce type des problèmes est abondamment étudié par des chercheurs, en appliquant aux réseaux sociaux ou systèmes de recommandation. Nous proposons une série d’algorithmes sur la base d’algorithme Passive-Aggressive pour résoudre des problèmes de Bandit contextuel. Nous profitons de sa fondations, et montrons que nos algorithmes sont plus simples à mettre en œuvre que les algorithmes en état de l’art. Ils réalisent des biens performances de classification. Pour des problèmes de Bandit Multi-objective (MOMAB), nous proposons une méthode motivée efficace et théoriquement à identifier le front de Pareto entre des bras. En particulier, nous montrons que nous pouvons trouver tous les éléments du front de Pareto avec un budget minimal dans le cadre de PAC borne. / Bandit problems constitute a sequential dynamic allocation problem. The pulling agent has to explore its environment (i.e. the arms) to gather information on the one hand, and it has to exploit the collected clues to increase its rewards on the other hand. How to adequately balance the exploration phase and the exploitation phase is the crux of bandit problems and most of the efforts devoted by the research community from this fields has focused on finding the right exploitation/exploration tradeoff. In this dissertation, we focus on investigating two specific bandit problems: the contextual bandit problems and the multi-objective bandit problems. This dissertation provides two contributions. The first contribution is about the classification under partial supervision, which we encode as a contextual bandit problem with side informa- tion. This kind of problem is heavily studied by researchers working on social networks and recommendation systems. We provide a series of algorithms to solve the Bandit feedback problem that pertain to the Passive-Aggressive family of algorithms. We take advantage of its grounded foundations and we are able to show that our algorithms are much simpler to implement than state-of-the-art algorithms for bandit with partial feedback, and they yet achieve better perfor- mances of classification. For multi-objective multi-armed bandit problem (MOMAB), we propose an effective and theoretically motivated method to identify the Pareto front of arms. We in particular show that we can find all elements of the Pareto front with a minimal budget.

Made available in DSpace on 2014-07-29T16:02:16Z (GMT). No. of bitstreams: 1 Dissertacao - Lays G C S de Jesus - Matematica.pdf: 936886 bytes, checksum: 303443d6b8eff2308a239c47a7c0d5af (MD5) Previous issue date: 2010-04-30 / In this work, we study the descent of methods for problem of optimization multiobjective which we introduce an order of relation induced by an closed convex cone.We study as it wiel calculate an descent of direction and we prove that every accumalation point of the sequence generated by the descent of methods with search of Armijo is weakly efficient. / Neste trabalho, estudamos o método de descida para problemas de otimização multiobjetivo, para o qual introduzimos uma relação de ordem induzida por um cone fechado e convexo. Estudamos como calcular uma direção de descida e provamos que todo ponto de acumulação da sequência gerada pelo método de descida com busca de Armijo é fracamente eficiente.

京都大学 / 新制・課程博士 / 博士(経済学) / 甲第24381号 / 経博第668号 / 新制||経||303(附属図書館) / 京都大学大学院経済学研究科経済学専攻 / (主査)教授 関口 格, 教授 原 千秋, 教授 NEWTON Jonathan Charles Scott / 学位規則第4条第1項該当 / Doctor of Economics / Kyoto University / DGAM

The use of multiobjective optimization in identifying systems that account for changes in customer needs, operating environments, system design concepts, and analysis models over time is generally not explored. Providing solutions that anticipate, account for, and allow for these changes over time is a significant challenge to manufacturers and design engineers. Products that adapt to these changes through the addition and/or subtraction of modules can reduce production costs through product commonality, and cater to customization and adaptation. In terms of identifying sets of non-dominated designs, these changes result in the concept of dynamic Pareto frontiers, or dynamic s-Pareto frontiers when sets of system concepts are simultaneously evaluated over time. In this dissertation, a five-step optimization-based design method identifying a set of optimal adaptive product designs that satisfy the predicted changes by moving from one location on the dynamic s-Pareto frontier to another through the addition of a module and/or through reconfiguration is developed. Development of this five-step method was separated into four phases. The first two phases of developments respectively focus on Pareto and s-Pareto cases, where changes in concepts, models, and environments that would effect the Pareto/s-Pareto frontier are ignored due to limitations in traditional optimization problem formulations. To overcome these limitations, and allow for these changes, the third phase of developments presents a generic optimization formulation capable of identifying a dynamic s-Pareto frontier, while the fourth phase adapts the phase three method to incorporate this new dynamic optimization formulation. Example implementations of the four phases of developments were respectively provided through the design of a modular UAV, a hurricane and flood resistant modular residential structure, a simple aircraft design example inspired by the Lockheed C-130 Hercules, and a modular truss system. Noting that modular products only represent one approach for dealing with changes in preferences, environments, models, and concepts, the final research contribution connects the presented method with parallel research developments in collaborative product design and design principles identification, followed by two case study implementations of this unifying design approach in the development of a modular irrigation pump and a modular plywood cart for developing countries.

This thesis addresses the challenges associated with thermal management systems (TMS) evaluation and selection in the conceptual design of hypersonic, air-breathing vehicles with sustained cruise. The proposed methodology identifies analysis tools and techniques which allow the proper investigation of the design space for various thermal management technologies. The design space exploration environment and alternative multi-objective decision making technique defined as Pareto-based Joint Probability Decision Making (PJPDM) is based on the approximation of 3-D Pareto frontiers and probabilistic technology effectiveness maps. These are generated through the evaluation of a Pareto Fitness function and Monte Carlo analysis. In contrast to Joint Probability Decision Making (JPDM), the proposed PJPDM technique does not require preemptive knowledge of weighting factors for competing objectives or goal constraints which can introduce bias into the final solution. Preemptive bias in a complex problem can degrade the overall capabilities of the final design. The implementation of PJPDM in this thesis eliminates the need for the numerical optimizer which is required with JPDM in order to improve upon a solution. In addition, a physics-based formulation is presented for the quantification of TMS safety effectiveness corresponding to debris impact/damage and how it can be applied towards risk mitigation. Lastly, a formulation loosely based on non-preemptive Goal Programming with equal weighted deviations is provided for the resolution of the inverse design space. This key step helps link vehicle capabilities to TMS technology subsystems in a top-down design approach. The methodology provides the designer more knowledge up front to help make proper engineering decisions and assumptions in the conceptual design phase regarding which technologies show greatest promise, and how to guide future technology research.

Manufacturers are increasingly being held responsible for the fate of their products once they reach their end-of-life phase. This research uses a combination of total disassembly time and recyclability to gauge the environmental impact of a product at this stage of its use. Recyclability, or wasted weight, is a function of the material contained by a product’s subassemblies as it is taken apart. This project suggests a graph-based method of representing product assemblies. Unlike many existing representation methods which are used in the field of automated disassembly, the method proposed here takes component connection methods into account. This, combined with a library of disassembly defining graph grammars, ensures that the disassembly simulation performed on this assembly approximates real-life disassembly procedures as closely as possible. The results of this simulation are Pareto sets whose contents represent various points in the disassembly process. Each member of the set is evaluated using the two primary parameters of disassembly time and wasted weight. This Pareto set can be used to judge a particular product’s performance during end-of-life, from the perspective of recyclability, against that of another product. / text

The areas of machine learning, data mining, and knowledge representation have many different formats used to represent information. Decision rules, amongst these formats, are the most expressive and easily-understood by humans. In this thesis, we use dynamic programming to design decision rules and analyze them. The use of dynamic programming allows us to work with decision rules in ways that were previously only possible for brute force methods. Our algorithms allow us to describe the set of all rules for a given decision table. Further, we can perform multi-stage optimization by repeatedly reducing this set to only contain rules that are optimal with respect to selected criteria. One way that we apply this study is to generate small systems with short rules by simulating a greedy algorithm for the set cover problem. We also compare maximum path lengths (depth) of deterministic and non-deterministic decision trees (a non-deterministic decision tree is effectively a complete system of decision rules) with regards to Boolean functions. Another area of advancement is the presentation of algorithms for constructing Pareto optimal points for rules and rule systems. This allows us to study the existence of “totally optimal” decision rules (rules that are simultaneously optimal with regards to multiple criteria). We also utilize Pareto optimal points to compare and rate greedy heuristics with regards to two criteria at once. Another application of Pareto optimal points is the study of trade-offs between cost and uncertainty which allows us to find reasonable systems of decision rules that strike a balance between length and accuracy.

Este trabalho apresenta o desenvolvimento de um procedimento de calibração para manipuladores industriais em ambiente virtual com o intuito de determinar os melhores valores de correção da cinemática direta para manipuladores industriais de até seis graus de liberdade. Para validação desta metodologia desenvolvida neste trabalho, o manipulador KUKA KR210 L100 2K foi escolhido como robô de estudo tanto para o sistema de simulação off-line quanto para a programação off-line. Para a determinação do valores de correção foi desenvolvido um algoritmo de calibração no ambiente Matlab que efetuar seis iterações contendo cada uma, um conjunto de doze valores de correção que estão divididos entre os ângulos "alpha" e "theta" dos parâmetros de Denavit-Hartenberg. Os dados de entrada para o algoritmo de calibração foram obtidos via Labview pelo controlador do KUKA KR210. Esses dados variam de acordo com a norma ISO 9283, que especifica que para aquisição de dados na calibração, é necessário variar a velocidade do robô em 10%, 50% e 100%, e onde cada trajeto de teste deve possuir trinta ciclos e um caminho contendo cinco pontos. Por último, para definir quais iterações são ideais para calibração, foi utilizado o método estatístico LSD de Fisher que tem a capacidade de mostrar quais variáveis cartesianas foram calibradas em relação a médias das posições nominais. Os resultados de calibração foram analisados pelo Diagrama de Pareto e pela Lei de Pareto 80/20 que identificaram as iterações mais adequadas que melhor expressam a equivalência mecânica e cinemática entre o robô real e o CAD do robô em ambiente off-line.

Turbo-machinery design is clearly about making many decisions often under uncertainty and with multiple conflicting objectives. In this work, a computational tool has been developed to assist in the preliminary design optimization of supersonic turbines with application on turbopump rocket engines. It was proposed an evolutive approach based on genetic algorithm to automatize the process of selecting values, based on hands-on experience, for decision variables and fulfill simultaneously decisive compromises faced by the designer. At design point, exploring multiple optima solutions, the tool allows to fast estimate in a robust and accurate manner, performance, main dimensions, mass of rotor-wheel and the lower possible flow rate of the turbine. Out of this point, performance maps can be calculated varying rotational speed and pressure ratio. Because, it is involved by many objectives, a Pareto-optimum set is found. The search ends when the relation power-to-weight converges. The power-to-weight ratio characterizes a good option to relate performance and weight among distinct turbopump turbines. The empirical loss models adopted as well as the recommended values in the selection of decision variables were obtained in the Russian literature. To demonstrate the functionalities of tool, the single-stage of the Soviet RD109 rocket turbine was redesigned. The results were validated from those found in the engine';s atlas of construction and also, by a statistical method of calculation defined from an experimental study that estimates the maximum turbine efficiency.