Thesis (M.A.)--Cincinnati Christian Seminary, 1987. / Abstract. Includes bibliographical references (leaves [80-81]).

In 2010 WikiLeaks released a number of secret and classified documents that contained information pertaining to the United States government. Since then, WikiLeaks and the United States government have been engaged in a rhetorical battle over the circulation of information. Using membership categorization analysis (MCA) as an analytical technique this thesis answers the following research question: what form(s) of politics are made possible as the result of the social orders produced by both WikiLeaks’ and the United States government’s public discourse on the circulation of information? After analyzing a sample of the related discourse, it became clear that the disagreement between WikiLeaks and the United States government is much greater than different views on the distribution of, and access to, information. Rather, the major issue is that the discourses produced by representatives of both organizations constitute two similar and yet somehow opposing social orders. The social orders produced result in different forms of politics and democracy. In turn, this involves each side thinking differently about transparency, the public, the government, the law, and the media.

The present study was calculated to determine whether or not a congregation with a good percentage increase in membership from a specific date in 1956 to a corresponding date in 1957 would also experience any percentage increase in attendance of members and visitors at the Sunday morning worship services. Likewise, would such a congregation have a large percentage of its assembly in attendance at its Bible classes and at its midweek services?

Submitted by Fabio Sobreira Campos da Costa (fabio.sobreira@ufpe.br) on 2017-06-22T14:09:03Z No. of bitstreams: 2 license_rdf: 811 bytes, checksum: e39d27027a6cc9cb039ad269a5db8e34 (MD5) Maria Rita Tese final pdf.pdf: 2723454 bytes, checksum: a5c4d5dde08d65f1641e2216cc5b4257 (MD5) / Made available in DSpace on 2017-06-22T14:09:04Z (GMT). No. of bitstreams: 2 license_rdf: 811 bytes, checksum: e39d27027a6cc9cb039ad269a5db8e34 (MD5) Maria Rita Tese final pdf.pdf: 2723454 bytes, checksum: a5c4d5dde08d65f1641e2216cc5b4257 (MD5) Previous issue date: 2016-05-19 / A tese reflete a interpretação adotada no Brasil, sobre o exercício da autonomia na constituição de filiação. Partindo de uma visão histórica do comportamento humano em família e a relativização dos conceitos e convenções no tempo, busca-se a justificativa para o limite binário do sistema de filiação que vem sendo criticado por crescente entendimento que o relativiza, em nome da ampliação da liberdade existencial. A socioafetividade e os avanços da biotecnologia na reprodução humana são frequentemente invocados pela doutrina e em decisões judiciais, que defendem a relativização da binariedade e a existência de novas “categorias” como a “multiparentalidade”. Tal relativização tornou-se uma tônica sem reflexões sobre o paradigma consequencialista, que é limitado em nosso sistema. A autonomia das relações privadas no Estado Liberal abstencionista pautou-se em um individualismo, que findou gerando sérios desequilíbrios na igualdade entre as partes. Já no Estado Social que intervém protetivamente, e busca garantir direitos fundamentais, a autonomia resvala de um interesse individual a um interesse coletivo e de ordem pública e submete o indivíduo a limites que visam a segurança e estabilidade nas relações existenciais. Nesse sentido, a tese conta com uma base doutrinária nacional e estrangeira, além de pesquisa jurisprudencial e sociojurídica na esfera da autonomia parental, contribuindo para uma revisão na aplicabilidade das normas de responsabilidade parental que proporcione soluções harmônicas em nosso sistema jurídico e garantia de sua unidade. / The thesis reflects the interpretation adopted in Brazil, on the exercise of autonomy in the constitution of membership. From a historical view of human behavior in the family and the relativization of the concepts and conventions in time, seeks the reason for the torque limit of the membership’system that has been criticized for growing understanding that relativize it, on behalf of the expansion of existential freedom. The partner affection and advances in biotechnology in human reproduction are often invoked by the doctrine and judicial decisions, defending the relativization of binarity and the existence of new "categories" like "multiparentality". Such relativism had become a tonic without reflections on the consequentialist paradigm, which is limited in our system. The autonomy of private relations in the Liberal State abstentionist was based on individualism, which ended causing serious imbalances in equality between the parties. In the welfare state intervening protectively, and seeks to ensure fundamental rights, autonomy slips of an individual interest to a public interest and public order and submits the individual limits that aimed at security and stability in existential relations. In this sense, the thesis has a national and foreign doctrinal basis, and jurisprudential and socio legal research in the sphere of parental autonomy, contributing to a review of the applicability of parental liability rules to provide harmonic solutions in our legal system and its unit warranty.

Thesis focuses on the issue of Georgia's possible receipt of Membership Action Plan (MAP), Thesis strives to represent the possibility of Georgia's MAP receipt and then subsequently in the geopolitics of NATO's enlargement. Main research areas of the thesis include the 1997. Second area of research is focused on the costs and benefits of Georgia's NATO gia's breakaway regions of Abkhazia and South

No abstract provided. / Thesis / Bachelor of Arts (BA)

No description available.

No description available.

No description available.

<p dir="ltr">Neural network-based machine learning models are now prevalent in our daily lives, from voice assistants~\cite{lopez2018alexa}, to image generation~\cite{ramesh2021zero} and chatbots (e.g., ChatGPT-4~\cite{openai2023gpt4}). These large neural networks are powerful but also raise serious security and privacy concerns, such as whether personal data used to train these models are leaked by these models. One way to understand and address this privacy concern is to study membership inference (MI) attacks and defenses~\cite{shokri2017membership,nasr2019comprehensive}. In MI attacks, an adversary seeks to infer if a given instance was part of the training data. We study the membership inference (MI) attack against classifiers, where the attacker's goal is to determine whether a data instance was used for training the classifier. Through systematic cataloging of existing MI attacks and extensive experimental evaluations of them, we find that a model's vulnerability to MI attacks is tightly related to the generalization gap---the difference between training accuracy and test accuracy. We then propose a defense against MI attacks that aims to close the gap by intentionally reduces the training accuracy. More specifically, the training process attempts to match the training and validation accuracies, by means of a new {\em set regularizer} using the Maximum Mean Discrepancy between the softmax output empirical distributions of the training and validation sets. Our experimental results show that combining this approach with another simple defense (mix-up training) significantly improves state-of-the-art defense against MI attacks, with minimal impact on testing accuracy. </p><p dir="ltr"><br></p><p dir="ltr">Furthermore, we considers the challenge of performing membership inference attacks in a federated learning setting ---for image classification--- where an adversary can only observe the communication between the central node and a single client (a passive white-box attack). Passive attacks are one of the hardest-to-detect attacks, since they can be performed without modifying how the behavior of the central server or its clients, and assumes {\em no access to private data instances}. The key insight of our method is empirically observing that, near parameters that generalize well in test, the gradient of large overparameterized neural network models statistically behave like high-dimensional independent isotropic random vectors. Using this insight, we devise two attacks that are often little impacted by existing and proposed defenses. Finally, we validated the hypothesis that our attack depends on the overparametrization by showing that increasing the level of overparametrization (without changing the neural network architecture) positively correlates with our attack effectiveness.</p><p dir="ltr">Finally, we observe that training instances have different degrees of vulnerability to MI attacks. Most instances will have low loss even when not included in training. For these instances, the model can fit them well without concerns of MI attacks. An effective defense only needs to (possibly implicitly) identify instances that are vulnerable to MI attacks and avoids overfitting them. A major challenge is how to achieve such an effect in an efficient training process. Leveraging two distinct recent advancements in representation learning: counterfactually-invariant representations and subspace learning methods, we introduce a novel Membership-Invariant Subspace Training (MIST) method to defend against MI attacks. MIST avoids overfitting the vulnerable instances without significant impact on other instances. We have conducted extensive experimental studies, comparing MIST with various other state-of-the-art (SOTA) MI defenses against several SOTA MI attacks. We find that MIST outperforms other defenses while resulting in minimal reduction in testing accuracy. </p><p dir="ltr"><br></p>