Published Article / Magnesium is the second most abundant intracellular cation with many control and regulatory functions. It regulates energy production and utilization and modulates activity of membrane ionic channels. Magnesium has direct control effects on cardiac myocyte ion channels making it useful in certain arrhythmias. Calcium is responsible for pacemaker excitation and for excitation-contraction coupling in myocytes but increased intracellular calcium produces early and late afterdepolarisations initiating arrhythmias. Magnesium regulates calcium channel activity preventing raised intracellular levels. Potassium channel activity is enhanced by magnesium hyperpolarizing the cell reducing arrhythmia generation. Magnesium is effective against long QT Torsade de Pointes. In rapid atrial fibrillation magnesium produces rate control slowing AV nodal conduction. Magnesium prevents digitalis toxicity due to associated hypomagnesemia.

No description available.

This thesis investigated the effects of intergroup contact on different types of collective action tendencies among advantaged and disadvantaged groups. Studies 1 and 2 tested the simultaneous effects of intergroup contact and established predictors of collective action on collective action tendencies and ingroup and outgroup oriented policies among Blacks and Whites in South Africa, and compared the effects of intergroup contact and social identity on collective action tendencies via relative deprivation and group efficacy. The findings revealed that while social identity was positively associated with collective action tendencies, both directly and indirectly, effects of contact were negative and indirect via relative deprivation and group efficacy. Studies 3 and 4 investigated the effects of contact and social identity on collective action tendencies via perceived threats. Using data from Turkish and Kurdish groups in Turkey, I found that social identity predicted collective action tendencies positively, both directly and indirectly, while it predicted outgroup attitudes negatively and indirectly via perceived threats. Intergroup contact, on the other hand, predicted outgroup attitudes positively, both directly and indirectly, and collective action tendencies negatively via perceived threats. In Study 5, intergroup contact was positively associated, both directly and indirectly, via perspective taking and collective guilt, associated with outgroup oriented collective action tendencies. In Study 6, the effect of social identity on ingroup oriented collective action was positive and direct. Intergroup contact with the weaker minority group, on the other hand, was positively associated with outgroup oriented collective action tendencies via perspective taking. Additionally, intergroup contact with the majority outgroup moderated this relationship. When participants reported more contact with the majority group, intergroup contact with the weaker minority was not associated with outgroup oriented collective action tendencies. However, when the participants reported less contact with the majority group, intergroup contact positively predicted outgroup oriented collective action tendencies. Finally, Study 7 investigated the effects of two different dimensions of contact, contact with the majority and minority on collective action, via outgroup attitudes, dual-identification, and common ingroup identity in a three wave longitudinal design (N=610) among Turkish Cypriots in northern Cyprus. While the results did not support findings from the previous studies on the so-called paradoxical effects of contact on collective action tendencies, they revealed a robust negative reciprocal relationship between outgroup attitudes toward Greek Cypriots and collective action tendencies.

published_or_final_version / abstract / Zoology / Master / Master of Philosophy

Collecting and annotating videos of realistic human actions is tedious, yet critical for training action recognition systems. We propose a method to actively request the most useful video annotations among a large set of unlabeled videos. Predicting the utility of annotating unlabeled video is not trivial, since any given clip may contain multiple actions of interest, and it need not be trimmed to temporal regions of interest. To deal with this problem, we propose a detection-based active learner to train action category models. We develop a voting-based framework to localize likely intervals of interest in an unlabeled clip, and use them to estimate the total reduction in uncertainty that annotating that clip would yield. On three datasets, we show our approach can learn accurate action detectors more efficiently than alternative active learning strategies that fail to accommodate the "untrimmed" nature of real video data. / text

Modular reinforcement learning is an approach to resolve the curse of dimensionality problem in traditional reinforcement learning. We design and implement a modular reinforcement learning algorithm, which is based on three major components: Markov decision process decomposition, module training, and global action selection. We define and formalize module class and module instance concepts in decomposition step. Under our framework of decomposition, we train each modules efficiently using SARSA($\lambda$) algorithm. Then we design, implement, test, and compare three action selection algorithms based on different heuristics: Module Combination, Module Selection, and Module Voting. For last two algorithms, we propose a method to calculate module weights efficiently, by using standard deviation of Q-values of each module. We show that Module Combination and Module Voting algorithms produce satisfactory performance in our test domain. / text

No description available.

No description available.

This study was undertaken to assess and describe a unique aspect of affirmative action planning in disciplines of higher education physical education. The two major purposes of the study were to (1) identify by sex, ethnic origin, and area of specialization faculty and doctoral students who were available for teaching and research in higher education disciplines of physical education, and (2) to develop a model for predicting future availability of faculty and doctoral students for teaching and research in higher education disciplines of physical education. The review of the literature indicated that there were a number of inflow and outflow parameters which must be considered in determining present availability and in predicting future availability. Research questions were formulated to address the academic workforce inflow and outflow parameters identified. Data were obtained from chairpersons of physical education departments at institutions offering advanced degrees in the disciplines of physical education and from the literature relating to the career patterns of doctoral recipients and to the academic workforce. Findings were based upon a descriptive analysis of the data with respect to each research question. A model for predicting current and future availability pools of all individuals for teaching and research in higher education disciplines of physical education was developed. Also, models were developed for predicting current and future availability pools of all white females and all male and female minorities for teaching and research in higher education disciplines of physical education. The models were based upon the inflow and outflow parameters as reported by the respondents to the questionnaire and in the literature. The currently predicted availability pool of individuals for teaching and research in higher education disciplines of physical education was found to be 3,160 faculty and doctoral students. The predicted availability pool of all individuals for teaching and research in higher education disciplines of physical education within the next two years was found to be 2,987 of the 3,448 faculty and doctoral students reported in the current availability pool.

Evidence Based Librarianship (EBL) seeks for and promotes the improvement of the librarian practice through the use of the best available evidence. Strongly used in medical contexts, Evidence Based Practice can be an important tool for the development of LIS, if practice is carefully thought and wisely combined with research and theoretical reflection. In order to achieve a state of equilibrium between theories and empirical studies, a qualitative method –action research- may be applied, as a dialogue between abstract ideas and the facts and signs provided by concrete experiences. Through action research, librarians can collect the evidence –using a series of qualitative tools- and use it for building theoretical knowledge in order to improve their work and their profession. From this viewpoint, after putting something into practice they will be able to know whether it worked as expected or not, make any change if it is necessary, and test the whole process again, searching more and new evidence. The method becomes a progressive helix that leads librarians to continuously evaluate their activities and services and improve them according of their final users’ needs. Fitting these ideas in the general context of “Library 2.0” new LIS model and in the particular situation of Social Sciences libraries, the conference briefly introduces some basic ideas on how action research should be employed for collecting and using evidence in LIS.