This research is an effort to improve the performance of successive approximation algorithm with a prime aim of solving finite states and actions, infinite horizon, stationary, discrete and discounted Markov Decision Processes (MDPs). Successive approximation is a simple and commonly used method to solve MDPs. Successive approximation often appears to be intractable for solving large scale MDPs due to its computational complexity. Action elimination, one of the techniques used to accelerate solving MDPs, reduces the problem size through identifying and eliminating sub-optimal actions. In some cases successive approximation is terminated when all actions but one per state are eliminated. The bounds on value functions are the key element in action elimination. New terms (action gain, action relative gain and action cumulative relative gain) were introduced to construct tighter bounds on the value functions and to propose an improved action elimination algorithm. When span semi-norm is used, we show numerically that the actual convergence of successive approximation is faster than the known theoretical rate. The absence of easy-to-compute bounds on the actual convergence rate motivated the current research to try a heuristic action elimination algorithm. The heuristic utilizes an estimated convergence rate in the span semi-norm to speed up action elimination. The algorithm demonstrated exceptional performance in terms of solution optimality and savings in computational time. Certain types of structured Markov processes are known to have monotone optimal policy. Two special action elimination algorithms are proposed in this research to accelerate successive approximation for these types of MDPs. The first algorithm uses the state space partitioning and prioritize iterate values updating in a way that maximizes temporary elimination of sub-optimal actions based on the policy monotonicity. The second algorithm is an improved version that includes permanent action elimination to improve the performance of the algorithm. The performance of the proposed algorithms are assessed and compared to that of other algorithms. The proposed algorithms demonstrated outstanding performance in terms of number of iterations and omputational time to converge.

Univ., Diss.--Hamburg, 2003.

Associé à la démocratie représentative, le vote en France est souvent observé à travers ses résultats (chiffre d’abstention, du vote blanc, inscription ou non inscription sur les listes électorales, etc.), ce qui laisse entendre que le fait démocratique serait caractérisé par les suffrages exprimés. Or, pour le citoyen le vote a un sens qui va au-delà du rituel des urnes. En tant que Sujet socio-psychologique, il vote en agissant autour d’enjeux fondamentaux dans le but de construire et de se construire par un acte puissant et utile en confirmant son histoire.Pour comprendre la signification du vote en tant qu’agir politique articulé à de nombreux agir (s) sociaux, eux-mêmes reposant sur des fondements psychologiques de l’agir humain, nous nous sommes positionnés en amont et en aval de la dynamique électorale et envisagé le comportement votant en fonction des connaissances, des croyances, des prédispositions normatives, affectives, bref, de l’attitude qui le sous-tend.La question générale de notre recherche est de décliner les conditions qui construisent le comportement votant. / Associated with representative democracy, voting in France is often seen through its results (abstention, blank vote, registration or non-registration on electoral lists, etc.), which suggests that the actual democratic would be characterized by the votes cast. For the citizen, the vote has a meaning that goes beyond the ritual of the ballot box. As a socio-psychological subject, he votes regarding fundamental issues in order to bring out solutions and construct himself through this historical action seen as a powerful, effective and useful. In order to understand the significance of voting as a political act expressed in many social actions, which are based on the psychological foundations of human action, we positioned ourselves upstream and downstream of the vote and considered the voting behavior relatively to knowledge, beliefs, normative, affective dimensions, in short, to the attitude that supports it.The general aim of our research is to specify the conditions that construct the voting behavior.

Thoughts about reflection and reflective practice have evolved over many decades, through carefully constructed theory and research applications, mainly based on the work of Dewey (1933) and Schön (1983). Evidence also exists in the literature that the ability to reflect on practice is considered a necessity for effective instruction (Sowder, 2007). By reflecting critically teachers become more positive in the search for a new understanding of their teaching practice and design more ways to deal with the challenges that confront them daily. When teachers act reflectively, they consider carefully the problems in their own teaching and think about how those problems are related to their educational or social context. They are aware of the consequences of their teaching and how their own assumptions or beliefs can influence their teaching. This main purpose of my research study was to explore the nature of mathematics teachers’ reflective practice in the context of lesson study. To achieve this aim, an in-depth exploration of five mathematics teachers’ reflection before, during and after teaching a lesson was conducted. The possible relationship between these teachers’ reflection and their classroom practice was also examined. The research also aimed to explore whether and how mathematics teachers’ reflections differ from the conceptualisations of reflection in classroom practice as found in the literature. Contextual factors that might influence the nature of mathematics teachers’ reflective practice were also investigated. My findings indicate that the mathematics teachers in my sample have a limited understanding of the concept of reflection. Furthermore, based on lesson plan analysis, there was no evidence that these teachers reflect-for-action. However, they all reflected on-action verbally and in writing, and three of the five teachers reflected-in-action while teaching. They all reflected on Level R1 (recall level of reflection) and Level R2 (rationalisation level of reflection) and three teachers reflected critically on their learners’ understanding of mathematics and their own teaching of concepts towards the end of the research project (Lee, 2005). Language and the lesson study group experience emerged as contextual factors that seemed to influence the teachers’ reflection. Although the research study’s results cannot be generalised due to the small sample, I believe that through engaging in the lesson study experience the five teachers of this study improved their reflective practice, reporting an increase in self-knowledge and finding new ways of teaching mathematics to learners. / Thesis (PhD)--University of Pretoria, 2011. / Science, Mathematics and Technology Education / unrestricted

No description available.

Static image action recognition aims at identifying the action performed in a given image. Most existing static image action recognition approaches use high-level cues present in the image such as objects, object human interaction, or human pose to better capture the action performed. Unlike images, videos have temporal information that greatly improves action recognition by resolving potential ambiguity. We propose to leverage a large amount of readily available unlabeled videos to transfer the temporal information from video domain to static image domain and hence improve static image action recognition. Specifically, We propose a video prediction model to predict the future video of a static image and use the future predicted video to improve static image action recognition. Our experimental results on four datasets validate that the idea of transferring the temporal information from videos to static images is promising, and can enhance static image action recognition performance. / Master of Science / Static image action recognition is the problem of identifying the action performed in a given image. Most existing approaches use the high-level cues present in the image like objects, object human interaction, or human pose to better capture the action performed. Unlike images, videos have temporal information that greatly improves action recognition. Looking at a static image of a man who is about to sit on a chair might be misunderstood as an image of a man who is standing from the chair. Because of the temporal information in videos, such ambiguity is not present. To transfer the temporal information and action features from video domain to static image domain and hence improve static image action recognition, we propose a model that learns a mapping from a static image to its future video by looking at a large number of existing images and their future videos. We then use this model to predict the future video of a static image to improve its action recognition. Our experimental results on four datasets show that the idea of transferring the temporal information from videos to static images is promising, and can enhance static image action recognition performance.

Mémoire numérisé par la Direction des bibliothèques de l'Université de Montréal.

Thesis with the topic "Types of action in The Civil Process" The objective of the thesis with the topic "Types of action in The Civil Process" is to offer a comprehensive view of the positive treatment of various types of action in the Czech civil procedural law. The action is a significant procedural institution, which is considered a universal means of protection of rights, and in both theory and practice it is treated with due importance. The thesis is divided in eleven chapters. In the first half, I define general characteristics of the action and its legal effects; in the second, I deal with the individual types of action in detail and analyse their specifics by means of a detailed of the literature and case law review method.

Contending with a changing climate presents a necessary push for planning. Although climate change is considered a global environmental problem requiring a global commitment and trans-national action, more and more, policymakers are recognizing the vital need for action at the local level. In the US, especially in the absence of national climate legislation, many local governments have begun developing strategic plans, or climate action plans (CAPs), to address adapting to impacts of climate change and mitigating greenhouse gas emissions. This thesis involves case studies of Portland, OR and Chicago, IL, cities with recently adopted CAPs and with considerable recognition in the field. The analysis involves an evaluation each city’s CAP and an evaluation of its implementation. The studies help elicit an understanding of the measures cites are employing to mitigate climate change and determine ways the planning profession can better assist communities in climate policy development and its prompt implementation.

The perception of an action and its production are inextricably linked. This entails that, during development, the skills that children are able to perform influence their perception of others\\\' actions. The present dissertation aimed to investigate the role of children’s experience on the perception of actions in three distinctive areas: manual actions performed by one person (individual action), manual actions performed by two people (joint action), and a conversation between two people. In order to succeed in each of the three areas, children have to acquire new skills and do so successively during their first three years of life. The methodological approach of this work was to measure the gaze behaviour of children, aged 6 months to 3 years, and adults during the observation of visually presented actions, which provided information on whether they were able to anticipate action goals. The findings obtained generally show an influence of experience on the anticipation of action goals in each of the three areas. First, a link between action and perception is not established as soon as an action emerges. There is at least some experience necessary for its development. Second, infants with no coordinated joint-action skills themselves anticipate the goals of joint action less well than those of individual action. Adults with considerable joint-action skills anticipate both equally well. And third, the course of a conversation can only be reliably anticipated by children aged 3 years and adults, whereas younger children shift their gaze between speakers randomly. Furthermore, only at the age of 3 years, did intonation support children’s anticipation of conversations.