Military conflicts have become nonlinear and the interrelated political and socio-economic changes within these conflicts have created new challenges for American policymakers. A tool called Wallis' Propositional Analysis (PA) suggests a new paradigm that includes thinking about complexity and robustness/systemicity in a policy. The purpose of this single case study was to determine how the PA paradigm adds heuristic value to complex policy decision-making. A backdrop of Wallerstein's complexity theory and complex adaptive systems (CAS) guided this study. This study examined policy statements from the Obama administration on the Afghanistan and Pakistan conflicts in late December 2009. Data were coded and analyzed using Wallis' specific methodological approach that includes a systematic analysis of the policy's propositions and complexity and robustness/systemicity. Key findings indicated that the PA paradigm offers a heuristic method for how to think about the interrelated propositions within a policy that reflect the expected changes the policy intends to make. Specifically, this study demonstrated that an interwoven PA structural approach to policymaking affords the policymaker a method to consider the complex and nonlinear changes in the policy environment. By applying the PA paradigm, policymakers can positively impact social change by exploring policy options that consider a range of possible outcomes from the policy proposal, prior to policy implementation.

Submitted by PPG Ci?ncia da Computa??o (ppgcc@pucrs.br) on 2018-10-04T19:48:28Z No. of bitstreams: 1 Paulo_Ricardo_Knob_dis.pdf: 15412485 bytes, checksum: d5fc868fa02ec06b2e29e6848c4b37ae (MD5) / Approved for entry into archive by Sheila Dias (sheila.dias@pucrs.br) on 2018-10-08T14:18:54Z (GMT) No. of bitstreams: 1 Paulo_Ricardo_Knob_dis.pdf: 15412485 bytes, checksum: d5fc868fa02ec06b2e29e6848c4b37ae (MD5) / Made available in DSpace on 2018-10-08T14:37:58Z (GMT). No. of bitstreams: 1 Paulo_Ricardo_Knob_dis.pdf: 15412485 bytes, checksum: d5fc868fa02ec06b2e29e6848c4b37ae (MD5) Previous issue date: 2018-08-22 / A maioria das t?cnicas dispon?veis hoje em dia para simula??o de multid?es est? focada em uma situa??o espec?fica, por exemplo, evacua??o em eventos perigosos. Poucas t?cnicas consideram os aspectos culturais e de personalidade presentes em uma sociedade para determinar o comportamento dos agentes. Portanto, este trabalho tem como objetivo construir um framework capaz de lidar com diferentes teorias culturais e de personalidade como input, traduzindo-as em uma parametriza??o de grupo, que vai determinar o comportamento de grupos e multid?es em ambientes virtuais. Al?m disso, inclu?mos no BioCrowds uma resposta de conforto para os agentes, em termos de densidade e caracter?sticas t?rmicas do ambiente. Os resultados indicam que os mapeamentos culturais / psicol?gicos parecem promissores, uma vez que os agentes foram capazes de se comportar conforme o esperado. Al?m disso, os agentes foram capazes de reagir devido ao conforto t?rmico e de densidade, melhorando sua capacidade de reagir ?s mudan?as do ambiente. / Most of the techniques available nowadays for crowd simulation are focused on a specific situation, e.g. evacuation in hazardous events. Very few of them consider the cultural and personality aspects present in a society to determine the behavior of agents. Therefore, this work aims to build a framework able to deal with different cultural and personality theories as input, and translate them into a group parametrization, which is going to determine the behavior of groups and crowds in virtual environments. Also, we include in BioCrowds a comfort response for agents, in terms of density and thermal characteristics of the environment. Results indicate that the cultural/psychological mappings seem promising, since agents were able to perform as intended. Additionally, agents were able to react due to thermal and density comfort, improving their ability to react to environmental changes.

Dans le cadre de cette thèse, nous avons mené trois études sur le terrain et/ou en laboratoire pour comparer l'importance relative de la géométrie des routes, représentée visuellement ou tactilement en deux dimensions, des informations multidimensionnelles extraites sur la base d'une visite directe du monde réel, et des informations symboliques indirectes sur les lieux par le biais d' instructions verbales, dans la construction de représentations spatiales chez l'homme, lui permettant de naviguer de mémoire dans des environnements complexes et non-familiers. Ces expériences ont permis de mettre en lumière certains aspects multidimensionnels et multi-sensoriels dans le traitement cognitif des informations spatiales et l'influence de celui-ci sur les performances d'hommes et de femmes, qui avaient pour tâche de retrouver de mémoire un itinéraire donné dans un milieu urbain non familier. Les résultats montrent clairement que des informations relatives aux repères visuels, transmises au moyen de séquences d'images panoramiques des itinéraires étudiés, sont inutilisables par un navigateur s'il ne dispose pas d'informations valides sur les distances relatives entre ces repères dans le monde réel (expérience 1). L'influence d'une exposition au préalable à un plan 2D visuel ou tactile des itinéraires a été comparé avec celle d'une expérience directe au moyen d'une visite guidée, ou indirecte au moyen d'indications verbales, sur les performances de navigation de personnes voyantes (expérience 1 et 2), déficientes visuelles (expérience 3), ou voyantes mais privées temporairement de leur vision (expérience 3) dans des environnements urbains à grande échelle. Les tests ont été réalisés en milieu réel (expériences 1 et 3) et virtuel (expérience 2) généré par ordinateur (Google Street View). Les performances ont été analysées en termes de temps du point de départ au point d'arrivée, nombre d'arrêts, nombre d'erreurs et taux de succès. Les stratégies potentiellement employées durant la navigation sont mis en avant sur la base des réponses des sujets à un questionnaire standardisé ; leurs capacités individuelles de se représenter l'environnement exploré sous forme d'images mentales a été évaluée sur la base de dessins. Les niveaux subjectifs de stress psychologique ont été mesurés pour mettre en évidences des différences possibles entre l'homme et la femme à cet égard. Les données ici montrent, d'une part, qu'une exploration rapide de représentations virtuelles correctement mises à l'échelle d'un environnement complexe permet aux sujets de retrouver cet itinéraire sans problème dans le milieu réel (expérience 2). Les personnes aveugles de naissance compensent l'absence de repères visuels dans la navigation efficacement par la mémorisation d'informations géométriques sur la base d'une brève exploration d'un plan tactile des itinéraires étudiés ici. Les sujets voyants privés de repères visuels, par contre, ne sont pas instantanément capables d'une telle compensation (expérience 3). Les résultats de ce travail sont discutés ici à la lumière des hypothèses actuelles sur la nature intrinsèque des représentations spatiales chez l'homme et placés ici dans le contexte d'un modèle de la mémoire de travail. Nous suggérons que cette dernière comprend des sous-systèmes multidimensionnels de stockage temporaire, capables de traiter en parallèle une multitude d'entrées sensorielles avec une capacité beaucoup plus grande que précédemment postulé dans le modèle classique de la mémoire de travail, qui présume un traitement sériel d'informations à capacité limitée. Un tel modèle est globalement mis en question par les résultats de cette thèse, qui ouvre une porte importante aux recherches futures sur le traitement cognitif d'informations spatiales chez l'homme dans un monde en perpétuel changement.Mots-clés : Environnements à grandes échelles – perception – traitement multi-sensoriel – représentation spatiale – navigation – humain / Previous studies investigating how humans build reliable spatial knowledge representations allowing them to find their way from one point to another in complex environments have been focused on comparing the relative importance of the two-dimensional visual geometry of routes and intersections, multi-dimensional data from direct exposure with the real world, or verbal symbols and/or instructions. This thesis sheds further light on the multi-dimensional and multi-sensorial aspects by investigating how the cognitive processing of spatial information derived from different sources of sensory and higher order input influences the performance of human observers who have to find their way from memory through complex and non-familiar real-world environments. Three experiments in large-scale urban environments of the real world, and in computer generated representations of these latter (Google Street View), were run to investigate the influence of prior exposure to 2D visual or tactile maps of an itinerary, compared with a single direct experience or verbal instructions, on navigation performances in sighted and/or visually deficient individuals, and in individuals temporarily deprived of vision. Performances were analyzed in terms of time from departure to destination, number of stops, number of wrong turns, and success rates. Potential strategies employed by individuals during navigation and mental mapping abilities were screened on the basis of questionnaires and drawing tests. Subjective levels of psychological stress (experiment 2) were measured to bring to the fore possible differences between men and women in this respect. The results of these experiments show that 2D visual maps, briefly explored prior to navigation, generate better navigation performances compared with poorly scaled virtual representations of a complex real-world environment (experiment 1), the best performances being produced by a single prior exposure to the real-world itinerary. However, brief familiarization with a reliably scaled virtual representation of a non-familiar real-world environment (Google Street View) not only generates optimal navigation in computer generated testing (virtual reality), but also produces better navigation performances when tested in the real-world environment and compared with prior exposure to 2D visual maps (experiment 2). Congenitally blind observers (experiment 3) who have to find their way from memory through a complex non-familiar urban environment perform swiftly and with considerable accuracy after exposure to a 2D tactile map of their itinerary. They are also able to draw a visual image of their itinerary on the basis of the 2D tactile map exposure. Other visually deficient or sighted but blindfolded individuals seem to have greater difficulty in finding their way again than congenitally blind people, regardless of the type of prior exposure to their test itinerary. The findings of this work here are discussed in the light of current hypotheses regarding the presumed intrinsic nature of human spatial representations, replaced herein within a context of working memory models. It is suggested that multi-dimensional temporary storage systems, capable of processing a multitude of sensory input in parallel and with a much larger general capacity than previously considered in terms of working memory limits, need to be taken into account for future research.Keywords: large scale environments – perception – multisensory processing - spatial representation – navigation - human

Previous studies investigating how humans build reliable spatial knowledge representations allowing them to find their way from one point to another in complex environments have been focused on comparing the relative importance of the two-dimensional visual geometry of routes and intersections, multi-dimensional data from direct exposure with the real world, or verbal symbols and/or instructions. This thesis sheds further light on the multi-dimensional and multi-sensorial aspects by investigating how the cognitive processing of spatial information derived from different sources of sensory and higher order input influences the performance of human observers who have to find their way from memory through complex and non-familiar real-world environments. Three experiments in large-scale urban environments of the real world, and in computer generated representations of these latter (Google Street View), were run to investigate the influence of prior exposure to 2D visual or tactile maps of an itinerary, compared with a single direct experience or verbal instructions, on navigation performances in sighted and/or visually deficient individuals, and in individuals temporarily deprived of vision. Performances were analyzed in terms of time from departure to destination, number of stops, number of wrong turns, and success rates. Potential strategies employed by individuals during navigation and mental mapping abilities were screened on the basis of questionnaires and drawing tests. Subjective levels of psychological stress (experiment 2) were measured to bring to the fore possible differences between men and women in this respect. The results of these experiments show that 2D visual maps, briefly explored prior to navigation, generate better navigation performances compared with poorly scaled virtual representations of a complex real-world environment (experiment 1), the best performances being produced by a single prior exposure to the real-world itinerary. However, brief familiarization with a reliably scaled virtual representation of a non-familiar real-world environment (Google Street View) not only generates optimal navigation in computer generated testing (virtual reality), but also produces better navigation performances when tested in the real-world environment and compared with prior exposure to 2D visual maps (experiment 2). Congenitally blind observers (experiment 3) who have to find their way from memory through a complex non-familiar urban environment perform swiftly and with considerable accuracy after exposure to a 2D tactile map of their itinerary. They are also able to draw a visual image of their itinerary on the basis of the 2D tactile map exposure. Other visually deficient or sighted but blindfolded individuals seem to have greater difficulty in finding their way again than congenitally blind people, regardless of the type of prior exposure to their test itinerary. The findings of this work here are discussed in the light of current hypotheses regarding the presumed intrinsic nature of human spatial representations, replaced herein within a context of working memory models. It is suggested that multi-dimensional temporary storage systems, capable of processing a multitude of sensory input in parallel and with a much larger general capacity than previously considered in terms of working memory limits, need to be taken into account for future research.

[pt] Esta dissertação de mestrado consiste em um estudo avançado sobre aprendizado profundo por reforço visual para robôs autônomos através de técnicas de transferência de aprendizado. Os ambientes de simulação testados neste estudo são ambientes realistas complexos onde o robô tinha como desafio aprender e transferir conhecimento em diferentes contextos para aproveitar a experiência de ambientes anteriores em ambientes futuros. Este tipo de abordagem, além de agregar conhecimento ao robô autônomo, diminui o número de épocas de treinamento do algoritmo, mesmo em ambientes complexos, justificando o uso de técnicas de transferência de aprendizado. / [en] This master s thesis consists of an advanced study on deep learning by visual reinforcement for autonomous robots through transfer learning techniques. The simulation environments tested in this study are highly realistic environments where the challenge of the robot was to learn and tranfer knowledge in different contexts to take advantage of the experiencia of previous environments in future environments. This type of approach besides adding knowledge to the autonomous robot reduces the number of training epochs the algorithm, even in complex environments, justifying the use of transfer learning techniques.

This study explores the application of multi-agent reinforcement learning (MARL) to enhance the decision-making, safety, and passenger comfort of Autonomous Vehicles (AVs)at uncontrolled intersections. The research aims to assess the potential of MARL in modeling multiple agents interacting within a shared environment, reflecting real-world situations where AVs interact with multiple actors. The findings suggest that AVs trained using aMARL approach with global experiences can better navigate intersection scenarios than AVs trained on local (individual) experiences. This capability is a critical precursor to achieving Level 5 autonomy, where vehicles are expected to manage all aspects of the driving task under all conditions. The research contributes to the ongoing discourse on enhancing autonomous vehicle technology through multi-agent reinforcement learning and informs the development of sophisticated training methodologies for autonomous driving.