Over the past few decades, we have been enjoying tremendous benefits thanks to the revolutionary advancement of computing systems, driven mainly by the remarkable semiconductor technology scaling and the increasingly complicated processor architecture. However, the exponentially increased transistor density has directly led to exponentially increased power consumption and dramatically elevated system temperature, which not only adversely impacts the system's cost, performance and reliability, but also increases the leakage and thus the overall power consumption. Today, the power and thermal issues have posed enormous challenges and threaten to slow down the continuous evolvement of computer technology. Effective power/thermal-aware design techniques are urgently demanded, at all design abstraction levels, from the circuit-level, the logic-level, to the architectural-level and the system-level. In this dissertation, we present our research efforts to employ real-time scheduling techniques to solve the resource-constrained power/thermal-aware, design-optimization problems. In our research, we developed a set of simple yet accurate system-level models to capture the processor's thermal dynamic as well as the interdependency of leakage power consumption, temperature, and supply voltage. Based on these models, we investigated the fundamental principles in power/thermal-aware scheduling, and developed real-time scheduling techniques targeting at a variety of design objectives, including peak temperature minimization, overall energy reduction, and performance maximization. The novelty of this work is that we integrate the cutting-edge research on power and thermal at the circuit and architectural-level into a set of accurate yet simplified system-level models, and are able to conduct system-level analysis and design based on these models. The theoretical study in this work serves as a solid foundation for the guidance of the power/thermal-aware scheduling algorithms development in practical computing systems.

The main task of this thesis is to clarify the impact of demographics on the rate of gross national savings which is considered one of the main elements of economic growth. Demographic trends and theoretical background are presented, which contributes to the understanding of the relationship of savings and the age structure of the population. The assessment of the relationship is done based on cross-sectional analysis of data from period 2009 -2013 for 133 countries, while the effect of demographic forces is studied through dependency rates of unproductive population. In the analysis partial estimates and analogous research with the pre-crisis data are performed. The results suggest that demographic forces leading to the growth of dependency rates negatively influence the savings rate. However, the important role is played by the level of development of individual countries or regional specifics since the effect of demographics between developed and developing countries is considerably different.

The South African primary rebate is governed by Section 6 of the Income Tax Act (58 of 1962). This primary tax rebate entitles taxpayers to a tax-free income portion up to a certain level depending on the rebate amount (also referred to as the tax threshold). The concept of tax thresholds in a tax system in essence adhere to the first tax canon of Smith (1776:676), which suggests that individuals should pay taxes in proportion to each person’s ability to pay tax. The implication of this tax canon is that individuals who have a limited or no ability to pay tax should only be subject to pay tax in relation to their ability. Therefore, before tax can be levied, an amount for the necessities-of-life must be deducted from the taxpayer’s income (Vivian, 2006:85). The primary rebate system thus gives individuals a tax-free income portion which is supposed to first compensate for an individual’s necessities-of-life expenses or put differently the costs to survive. The main purpose of the present study is to critically analyse and compare the fairness of the primary rebates in South Africa in relation to other countries. To meet the main purpose a comparison was done between South Africa’s primary rebate and related government grant programs to that of Brazil and Australia. It was found that South Africa rebate system creates significant vertical and horizontal unfairness and that it compares poorly to the fairer multiple rebate and government grant systems of Brazil and Australia. Accordingly it was recommended that the unified primary rebate system of South Africa is reviewed and brought in line with the multiple rebate systems implemented in countries such as Brazil and Australia. / Dissertation (MCom)--University of Pretoria, 2012. / Taxation / unrestricted

No description available.

Cross-lingual transfer has been shown effective for dependency parsing of some low-resource languages. It typically requires closely related high-resource languages. Pre-trained deep language models significantly improve model performance in cross-lingual tasks. We evaluate cross-lingual model transfer on parsing Marathi, a low-resource language that does not have a closely related highresource language. In addition, we investigate monolingual modeling for comparison. We experiment with two state-of-the-art language models: mBERT and XLM-R. Our experimental results illustrate that the cross-lingual model transfer approach still holds with distantly related source languages, and models benefit most from XLM-R. We also evaluate the impact of multi-task learning by training all UD tasks simultaneously and find that it yields mixed results for dependency parsing and degrades the transfer performance of the best performing source language Ancient Greek.

No description available.

This thesis examines the way in which China's share of the rare earths global industry is constructed as a threat. The central argument of the research is that the US discourse presented a new form of danger through the lens of classical geopolitical thinking, thus, creating a new type of prospective conflict derived from resources that are perceived to be scarce and yet relatively abundant. It argues that the construction of this 'threat' that the US faces is a constant articulation of perceived vulnerabilities in shaping geopolitical identities and reinforcing ideologies through which are carried out by different actors. Rare earths are crucial for modern conditions and their applications include commercial, military and green technology but mining and production are neither economical nor environmentally friendly. The People's Republic of China is the principal exporter of these rare metals, but because of a territorial dispute in the South China Sea in 2010 it has been labelled as a monopoly power. This study seeks to examine the constructed threat that China poses to the US. China, here, is not only constructed to be a resource hungry giant but also a malicious state that would utilise its 'monopoly' status as geopolitical leverage. The thesis will employ discourse analysis and wisdoms of...

Multi-label learning addresses the problem that one instance can be associated with multiple labels simultaneously. Understanding and exploiting the Label Dependency (LD) is well accepted as the key to build high-performance multi-label classifiers, i.e., classifiers having abilities including but not limited to generalizing well on clean data and being robust under evasion attack. From the perspective of generalization on clean data, previous works have proved the advantage of exploiting LD in multi-label classification. To further verify the positive role of LD in multi-label classification and address previous limitations, we originally propose an approach named Prototypical Networks for Multi- Label Learning (PNML). Specially, PNML addresses multi-label classification from the angle of estimating the positive and negative class distribution of each label in a shared nonlinear embedding space. PNML achieves the State-Of-The-Art (SOTA) classification performance on clean data. From the perspective of robustness under evasion attack, as a pioneer, we firstly define the attackability of an multi-label classifier as the expected maximum number of flipped decision outputs by injecting budgeted perturbations to the feature distribution of data. Denote the attackability of a multi-label classifier as C∗, and the empirical evaluation of C∗ is an NP-hard problem. We thus develop a method named Greedy Attack Space Exploration (GASE) to estimate C∗ efficiently. More interestingly, we derive an information-theoretic upper bound for the adversarial risk faced by multi-label classifiers. The bound unveils the key factors determining the attackability of multi-label classifiers and points out the negative role of LD in multi-label classifiers’ adversarial robustness, i.e. LD helps the transfer of attack across labels, which makes multi-label classifiers more attackable. One step forward, inspired by the derived bound, we propose a Soft Attackability Estimator (SAE) and further develop Adversarial Robust Multi-label learning with regularized SAE (ARM-SAE) to improve the adversarial robustness of multi-label classifiers. This work gives a more comprehensive understanding of LD in multi-label learning. The exploiting of LD should be encouraged since its positive role in models’ generalization on clean data, but be restricted because of its negative role in models’ adversarial robustness.

This study aimed to examine the possible intrapersonal (i.e., attention, depression) and interpersonal (i.e., social relationships with friends, social relationships with teachers) problems related to mobile phone dependency and their impacts on academic achievement in adolescents in South Korea using a national sample of 2159 middle and high school students (1074 male and 1085 female). A structural equation modeling approach with mediation analysis was employed to test the seven hypotheses drawn from conceptual and empirical bases. Results showed that mobile phone dependency negatively predicted attention and positively predicted depression, which in turn, affect social relationships with friends and both Korean language arts and mathematics achievement. Also, the mediating roles of attention, depression, and relationships with friends were found between mobile phone dependency and the academic achievement of middle and high school students in S. Korea. Based on the findings, implications of the current study and future directions for research were discussed.

A child development laboratory in a university finds its first justification in its ability to contribute to the education of college students. After this justification has been established, the laboratory has a responsibility for serving the children enrolled in it.