AUB ScholarWorks (American Univ. of Beirut)
Not a member yet
33634 research outputs found
Sort by
Infering Underlying Networks from Time Series of Dynamical Systems and Evaluating Global Balance
Get PDFA complex system’s emerging behavior is a result of the interactions of its components. A graph-theoretic representation of it is a network of interactions dictating
through differential equations the evolution of the state of the individual components, represented by nodes. These networks can be signed, directed, and weighted.
Our first goal is to infer these networks of interactions from time series relying on
dynamical systems theory. Our second goal is to characterize these networks, and
for this purpose, we rely on multiscale definitions of the frustration indices. We
implement algorithms that compute the indices of frustration on multiple levels, explore and address some of the computational bottlenecks, and apply the algorithms
to the network inferred from the dynamics
Unique and You
No full textA children's book, created by AUB students from the Education Department, for the course EDUC218 as a final project.First day of school butterflies? No problem! James meets a vibrant cast on his first school day, each with something special. Unique and You celebrates diversity and self-acceptance, reminding young readers to embrace what makes them extraordinary
An Inflationary Scenario in Mimetic Gravity
Get PDFWe study the Starobinsky potential model of inflation in mimetic gravity where the box of the mimetic field (d’Alembertian) is coupled to the inflation potential, and then note that the observable quantities, namely the tensor-to-scalar ratio in perturbations and the spectral index agree with observations. To that end, a review of cosmology and inflation in ordinary General Relativity is presented. The results of Higgs inflationary scenario and Starobinsky/R2 model are noted, specifically the tensor-to-scalar ratio and the spectral index which are proportional to the inverse
square of the number of e-folds N and to the inverse of N respectively during inflation. Then Mimetic gravity, a modified theory of gravity which incorporates dark matter in the underlying geometry of spacetime is revisited. This incorporation is due to writing the physical metric as an auxiliary metric multiplied by a conformal mode. We then proceed to setup the theory of mimetic inflation by coupling a linear function of the d’Alembertian of the mimetic field with the inflaton potential in the action. We get the equations of motion, analyze them in the modified slow roll approximation, and show that an inflationary epoch takes place. After proving that a reheating phase indeed occurs in our model, we then carry cosmological perturbations to the background equations of motion to get the scalar and tensor power spectrum. It is shown afterwards that for our Starobinsky potential model the observable quantities (spectral index and tensor-to-scalar ratio) agree with observations and with Higgs inflationary scenario / R2 inflation. Finally, a very brief look on self-reproduction/eternal inflation and how to possibly avoid it based on this work is mentioned
Investigating the Role of miR-187-3p in Molecular Apocrine Breast Cancer
Get PDF: Breast cancer (BC) is a health problem that is reported to have the highest incidence and mortality rates of cancer among females worldwide and in Lebanon. Moreover, BC heterogeneity renders it from being diagnosed or treated as a single entity. As such, there has been several attempts to classify BC. The expression of hormonal receptors is one of the methods being used for subtyping BC. These receptors are estrogen receptor (ER), progesterone receptor (PR), and human epidermal growth factor receptor 2 (HER2). Thus, based on the expression of these receptors, BC is classified into 4 molecular subtypes. The most aggressive subtype is negative for the three receptors and is called triple negative BC (TNBC). TNBC itself is also heterogenous and it overlaps with distinct subtypes, namely Molecular Apocrine (MA) and Luminal Androgen (LAR) BCs. These subtypes which overexpress Androgen receptor (AR) are, even today, still misdiagnosed and poorly treated. The discovery of new biomarkers that can help in differentiating subtypes might serve as a helpful tool. A significant amount of research has proved these micro-RNAs (miRNAs) are dysregulated in different types of cancers including BC, and consequently could be involved in cancer initiation and/or progression. Moreover, some miRNAs exhibit distinct dysregulation patterns across BC subtypes, offering potential diagnostic value. Using publicly available data and in silico analysis done on a cohort of ER-negative tissues, our lab found that miR-187-3p is significantly upregulated in samples classified as MABC compared to non-MABC samples. In addition, miR-187-3p is reported as a key player in the tumorigenesis of several types of cancer by controlling proliferation, invasiveness, migration, and aggressiveness of cancerous cells. Building upon these findings, we investigated the role of miR-187-3p in the tumorigenesis of MABC using in vitro BC cellular models
The Role of Governance in Enabling and Sustaining School-Based Improvement: The TAMAM Project Experience with the Lebanon Hub
Get PDFThis study explored the role that governance plays in enabling and sustaining school-based improvement through the TAMAM project experience with six public schools which have been partnered with it since 2015 and which constitute part of its hub in Lebanon. It aimed to find out which governance components have challenged or facilitated the selected schools’ organizational performance, particularly the implementation of their improvement projects and staff’s readiness to innovate. The study adopted a qualitative methodology and collected data mainly through the analysis of documents. This documented data was of a wide variety and comprised transcripts of raw data, memos from interviews, project progress reports, and technical reports. The analyzed data fell into two main categories: those primarily accessed through a subset of the TAMAM project databank and secondly those made publicly accessible on governmental websites in the form of legislative decrees.Following this, I analyzed data using mainly a deductive process with predetermined codes developed from a conceptual framework which I synthesized from the lessons learnt of prominent researchers’ empirical studies about how we can sustain school improvement through governance as well as studies which explored practical applications of a decentralized governance model known as the school-based management model. I then sorted this data into these generated codes and made a comparison between the two to form an in-depth understanding of the studied phenomenon. As a final step, I left room for some form of inductive analysis to occur and alternative themes to emerge. The study found three governance components which facilitated schools’ implementation of their improvement projects in the context of the study. Namely, these were the established trust and positive informal communication channels with governmental supervisors; the informal mentorship that school principals received from officials versed in the governance structure and system functioning; and the municipal financial support. As for the governance components which were found to hinder implementation, the study found seven of them. The first three were the absence of formalized two-way, consistent communication with governmental supervisors; the lack of municipal coordination in support of schools’ improvement initiatives; and the lack of clear communication and sufficient information provided by the ministry about mandated educational policies and initiatives. The remaining four were overloading schools with top-down projects which were not aligned with their own initiatives; the centralized decision-making process in the ministry; the mismanagement in the task allocation of schools’ human resources; and the absence of regular teacher training opportunities which are responsive to the needs of teachers. As for the second part of the study’s research question exploring the effect of governance on school lead team’s readiness to innovate, the study only found derailing governance components. Namely, these were the presence of an outdated, restrictive mandated curriculum; the absence of mentors limiting teachers’ ability to generate innovative ideas; prescribed professional development that does not cater to teachers’ personal aspirations; lack of sustainable resources for teachers’ engagement in innovation; and lack of emphasis on professional qualification in teacher recruitment. Finally, the study discussed the results through a comparative lens with the proposed conceptual framework and the literature to help generate informed recommendations for future research and practice. One recommendation for future research was to conduct a comparative study between schools which are partnered with TAMAM and public schools whose practitioners have only taken part in top-down improvement initiatives mandated by the ministry or schools whose practitioners have yet to acquire the needed competencies to lead and enact school-based improvement. As for the recommendations for practice, one of the study’s main recommendations was to devolve more decision-making power to regional education office staff members so that educational decisions become informed by schools’ needs and are responsive and timely to requests submitted by schools
Experiences of Maronite Membership - An Ethnographic Study of Family and Community Integration
No full textThis thesis seeks to unpack sentiments of belonging and exclusion within the parameters of Lebanese Christian families and community. As a result, topics within the themes of kinship, inclusion, responsibility, security, religious identity, and solidarity are explored. The objective of this study is to explore whether instances of racism, xenophobia, and sexism are a necessary condition or accidental circumstance of Lebanese Christian community preservation and development. With this objective in mind, this thesis aims to understand (1) the relationship between sectarian identity and family expectations, upbringing, and morals (2) the spectrum of unconscious to deliberate processes of othering (3) the threat and fear associated with exclusion from belonging to Christian communities, as well as (4) the effects of rejecting social norms and family expectations. The data presented in this thesis is based on ethnographic fieldwork conducted sporadically between the years 2020 to 2024. As a result, this thesis presents a nuanced perspective of the COVID19 pandemic and lockdown period, economic crisis and nationwide electricity shortages, as well as the emergence of right-wing alternative security groups in the historically Christian-majority neighborhood of Ashrafieh in Beirut
A Portrait of Jnoud El Rab: A Case of Religious Extremism in Lebanon
No full textIn the past three years, Lebanon have witnessed the emergence of Jnoud El Rab, a group portrayed by media and some individuals as an extremist Christian group due to their actions and statements against Syrian and Palestinian refugees, and individuals who do not follow the true Christian faith, and other groups like the LGBTQ community. The recent literature shows that religious extremism is a multi-layered perpetuated by political, global, social, individual, and religious levels. Religious extremism in Lebanon has been studied through related to the state’s weakness and salient sectarian identities. There has not been a comprehensive recent literature on the emergence of Christian extremism in Lebanon or on the emergence of Jnoud El Rab. Therefore, in my thesis, I attempted to draw a portrait of Jnoud El Rab and identified the factors that qualify them as a religious extremist group based on the literature. In my research questions, I have conducted in depth interviews with two members of the group as well as undertaken a discourse analysis of the content on their Facebook and YouTube pages. Studying Jnoud El Rab and situating them into the literature of religious extremism will not only allow us to improve our understanding of the dynamics of the group but also to broaden our understanding of Christian religious extremism in a country like Lebano
FedSAM: Sharpness-Aware Minimization for Improved Generalization Under FL Settings
No full textWhile being extensively studied in machine learning community, the problem of improving generalization in Federated Learning (FL) is still in its infancy. The main challenge stems from the heterogeneous nature of client data and the varying computational capacity of clients. Many researchers have recently linked the generalization gap to the sharpness of the landscape of the optimization model. In Foret P. et al, Sun Y. et al, and Qu Z. et al, a Sharpness-Aware Minimization (SAM) framework that seeks flat minima by penalizing sharp regions was introduced. In this thesis, we propose a SAM-like approach for improving generalization in FL settings. Unlike several existing methods that
incorporate SAM when training local models, our proposed framework penalizes the loss of the global function. To motivate our approach, we first provide a counter-example that shows that finding flat minima for local clients does not necessarily result in a flat aggregation for the global model. Furthermore, we develop an efficient sharpness-aware algorithm that adaptively computes global gradient similarity parameters for penalizing sharp regions. Harnessing these similarity parameters, a distinct sharpness penalty parameter is shared with each client. In particular, clients with varying local data distribution receive different penalty terms. We mathematically established the convergence of our suggested algorithm. Then, to demonstrate the efficiency of our algorithm, we perform several experiments on MNIST, FMNIST, and CIFAR datasets. Our results show a significant increase in generalization performance compared to existing approaches
Data Science for Lebanon Public Policy
Get PDFThis policy brief examines the impact of data science on the public sector in the Middle East and North Africa (MENA) region in general, with a particular focus on Lebanon, based on insights from discussions and technical talks of invited data science experts at the 2023 Women in Data Science (WiDS) conference at the American University of Beirut (AUB),
held on April 27-28, 2023, at the Suliman S. Olayan School of Business at AUB. The report also utilizes existing secondary data on the current situation of data science in Lebanon specifically and the MENA region more broadly. The brief starts with an overview of the importance of data science in decision-making. Then, it presents a review of data science in public sectors across the MENA region, with emphasis on key success stories of how data science has improved the public sector in several MENA countries. The brief then focuses on the case of Lebanon, highlighting how
data science can be used to improve the public sector in Lebanon, provide better disaster preparedness, and help mitigate the financial and economic crises in the country. The brief concludes with policy recommendations and interventions on how Lebanon can harness the potential of data science to drive improvements in governance, policy formulation, disaster management, and public service delivery, ultimately enhancing the overall performance of the public sector