Alliance One Tobacco (Malawi)
Academic Research Repository at the Institute of Developing EconomiesNot a member yet
52305 research outputs found
Sort by
The effects of trait-associated variation on transcription factor binding and other regulatory phenotypes
No full textUnderstanding the genetic causes of disease is one of the fundamental goals of human genetics. Disease variants can cause dysregulation of genes, either by altering the nucleotide sequence of regulatory elements or by modifying the amino acid sequence of regulators, such as transcription factors. A widely applied strategy to search for genetic loci linked to common diseases and complex traits is genome-wide association studies (GWAS). However, genetic maps derived from GWAS implicate genomic loci with a set of correlated (i.e., virtually indistinguishable based on statistics) variants. Hence, the identities of the causal variants and causal genes are key questions that remain. In Chapters 2 and 3, I address this challenge of elucidating gene regulatory mechanisms in noncoding GWAS loci. In Chapter 2, I present a strategy of harnessing transcription factor occupancy variation data to pinpoint putative causal variants at trait-associated loci. In Chapter 3, I explore whether chromatin accessibility variation sheds light on why disease-associated gene expression variation is missing in many loci. For research of more severe diseases, such as structural birth defects, family trio studies are commonly conducted. In Chapter 4, I search for genes associated with structural birth defects by combining genetic data from multiple family trio cohorts, with the aim of identifying candidate disease variants in transcription factor genes. In sum, this dissertation discusses the challenges of pinpointing disease-causing variants and genes and presents possible strategies to address them.Medical SciencesMedical Science
An area law for 2d frustration-free spin systems
No full textWe prove that the entanglement entropy of the ground state of a locally gapped frustration-free 2D lattice spin system satisfies an area law with respect to a vertical bipartition of the lattice into left and right regions. We first establish that the ground state projector of any locally gapped frustration-free 1D spin system can be approximated to within error ϵ by a degree O(nlog(ϵ−1)‾‾‾‾‾‾‾‾‾√) multivariate polynomial in the interaction terms of the Hamiltonian. This generalizes the optimal bound on the approximate degree of the boolean AND function, which corresponds to the special case of commuting Hamiltonian terms. For 2D spin systems we then construct an approximate ground state projector (AGSP) that employs the optimal 1D approximation in the vicinity of the boundary of the bipartition of interest. This AGSP has sufficiently low entanglement and error to establish the area law using a known technique.Accepted Manuscrip
Accelerating Atomistic Simulations with GPUs and Machine Learning
No full textThe field of atomistic simulations is rapidly evolving. The simulations are becoming more reliable and accurate through the invention of ever more advanced ML potentials such as uncertainty aware and equivariant methods. Simultaneously, the leading GPU-based supercomputers are finally reaching the exascale, which paves the way for larger and longer simulations. This thesis revolves around the intersection of these advances and the acceleration of state-of-the-art ML potentials with modern GPUs on the world's largest supercomputers.
More concretely, my focus has been the performance portability and scalability of FLARE and Allegro, two ML potentials near the opposite ends of the cost-accuracy trade-off. FLARE is a sparse Gaussian process potential that attempts to push the envelope for speed while maintaining reasonable accuracy. I have developed a Kokkos implementation of FLARE that outperformed previous state-of-the-art methods by 70\% on the second fastest supercomputer in the world, and recently reached one trillion atoms on Frontier, the fastest supercomputer. Allegro is an equivariant neural network implemented in PyTorch which sacrifices some speed to reach leading accuracy while maintaining scalability through its innovative architecture. Through the scalability and my LAMMPS interface, Allegro was able to efficiently utilize 5120 GPUs and reach relevant speeds for a wide range of biomoleular structures
The Distinction between Theology and Ethics: A Critical History
No full textABSTRACTThis article sketches an intellectual history of the distinction between Christian theology and Christian ethics. The twists and turns of that history have been obscured by a recent tendency to deny the distinction's usefulness, as part of a wider strategy for reasserting theology's relevance to modern social problems. By contrast, earlier theologians assumed the value of the theology/ethics divide, interpreting it through Aristotelian, neo‐Kantian, and finally Marxist categories. The distinction fell into disrepute because theologians struggled to maintain the distinction consistently and disagreed on the concerns implicated by it, variously using it to affirm the moral subject's agency, the divine/human difference, or the complexity of real people's circumstances. Nonetheless, the distinction has persisted as a useful shorthand for recognizing the limitations of Christian theology, qua a conceptual discourse, in fully apprehending its subject matter of the Christian life.Version of Recor
On Pourroit les Multiplier Presque Jusqu’au Nombre des Paroisses: A Study of the Dialects in Middle Breton
No full textThis dissertation approaches the question of the existence of dialectal variation in Middle Breton by conducting a systematic study of the language found in the literature from that period. Middle Breton is a language that was used from the twelfth century to 1659, and was long thought to have had no dialects. I demonstrate not only that dialects existed in this stage of the language, but also that traces of them can be found, in spite of the use of a written standard.
The first chapter surveyed the existing literature, discussing past and present theories regarding the existence of dialects in Middle Breton. It concluded that dialects must have existed in Middle Breton as they do in all languages throughout the world. The surviving Middle Breton texts show, however, a relatively unified form of the language. The purpose of this study was to examine Middle Breton texts to see if any dialect traces could be perceived through the uniform spelling.
The second chapter analysed documents of the Middle and Early Modern Breton period discussing dialects. These shed light on the perception of the language and its dialects. Its conclusion was that people were aware of dialectal variation, and that there were four dialects, Leon, Kerne, Treger, Gwened, the same four found in Early Modern and Modern Breton. These appear to have also followed the diocese borders, as they do in Early Modern and Modern Breton.
The third chapter first discussed the validity of using Early Modern dictionaries and modern dialect studies to analyse older, Middle Breton, material. I thus present each dialectal feature found in Middle Breton in turn, showing evidence from Early Modern and Modern sources for its dialectal spread and variation. I then examine each Early Middle Breton text in turn. This showed that dialectal features existed in written Early Middle Breton. They were not common, but were the same features found in Classical and Late Middle Breton texts, as well as in Early Modern Breton texts.
The fourth chapter looked at the Classical Middle Breton texts and examined them in turn, based on the dialectal baseline established in Chapter Three. I found that there were plenty of dialectal features in each text, though expectedly more in longer texts. These features were dialectally consistent throughout the texts and matched well with the modern dialect maps. I also found that a remarkable proportion of the texts came from the same area: the city of Morlaix. When reading the texts from this period, it gives the impression that they are very similar in language, though not necessarily because of a standardized language.
The fifth chapter looked at the Late Middle Breton texts. I found that the texts in this chapter contained more dialect features than the earlier ones, though the same ones appearing more often, and more consistently. Late MB texts also came primarily from the area around Morlaix giving the impression that the language was quite standardized, with a few odd texts: those that were not from Morlaix
Twisting History for Hate: Nordicism, Norse Pseudohistory, and Twenty-First Century American White Nationalism
No full textNordicism—the belief in a Nordic or Aryan “master race”—emerged in nineteenth century anthropology as a form of scientific racism. As this ideology spread, a romanticized pseudohistorical version of Northern Europe flourished—most noticeably during Germany in the 1930s and 1940s. It continues to exist in the minds of white supremacists, neo-Nazis, the alt-right, and white nationalists today, with hate crimes and racial prejudice headlining the news in the United States daily. Understanding the patterns many extremists show on social media and internet forums is crucial—especially in the context of historical movements and ideas. This historiographical and phenomenological study presents a holistic view of the history and evolution of Nordicism and Norse pseudohistory and their modern use by white nationalists in the twenty-first century United States. It covers a variety of sources that speak on white nationalism, scientific racism, eugenics, and the fantastical concept of “Northernness.” Collecting data from self-identified white nationalists and white supremacists on social media and internet forums, as well as incidents of racially motivated hate crimes, can help answer the question of how modern white nationalists use debunked science and history to promote their ideology. Results of the research showed a clear connection to Nordicism and Norse pseudohistory while also admitting a lack of direct racist content in many examples. This thesis aims to help identify those who would commit racially biased crimes, as well as fight disinformation
Essays in Behavioral and Experimental Economics
No full textThis dissertation consists of essays in behavioral and experimental economics, with a focus on information processing and its economic implication.
The first chapter, coauthored with Yuen Ho, studies the Spiral of Silence theory. According to this idea, people who hold perceived socially inappropriate views self-censor, generating a distribution of expressed views that is skewed towards appropriate opinions. If the attention paid to silence is limited, this can exacerbate self-censorship and create an equilibrium where only socially appropriate views are expressed and considered dominant. We experimentally test this hypothesis based on a simple model in which self-censorship and limited attention to silence interact to jointly establish equilibrium norms. In our experiment, UC Berkeley undergraduates discuss controversial political and socioeconomic issues. Students with socially inappropriate views self-censor to a significant degree. Given the limited attention students pay to silence, self-censorship amplifies over time. We experimentally increase the salience of silence and show that this affects both beliefs about others’ views and public expression decisions. Because inference and expression amplify each other, different levels of attention to silence can produce divergent perceived social norms in equilibrium.
The second chapter, co-authored with Yixi Jiang and Ziqi, explores correlation neglect on social media and its economic implication. In a field experiment on a popular Chinese social media site, we exposed Chinese college students to redundant negative posts about civil service jobs. Consequently, they developed more negative view of these jobs and became less likely to register for civil service exams, compared to the control group. Our study demonstrates that correlation neglect can affect beliefs and real-life outcomes, including career choice.
The third chapter, co-authored with Juanjuan Meng and Xi Weng, studies how awareness of media bias affects news consumers' reactions to potentially biased news. We propose that news consumers’ tendency to repost news articles whose political inclination is inconsistent with the media outlet's general ideology, defined as “inconsistent news”, is an indicator of their strategic responses to news. By combining data from Chinese microblog users and an online experiment that exogenously varies whether news sources are revealed, we provide both observational and causal evidence that Microblog users are more likely to repost inconsistent news, an indication of strategic reaction to media bias. We also show that simply reminding people of media bias can raise their awareness and change their reactions to news reports on social media
Operational Energy Strategy: Turning the U.S. Military into the most Energy-Efficient Fighting Forces in the World
No full textThe purpose of this paper is to identify, develop, improve, and implement a series of options in operational energy strategy for the Department of Defense utilizing renewable energy sources and turning the U.S. Military into the most Energy-Efficient Fighting Forces in the World (Coren). Energy is an important source of America’s progress, prosperity and development but also it is the basis of instability, crisis, inequality and prominently, environmental deterioration and risk. All those factors are highly accentuated in our military operations where every combatant command has a significant energy related mission. Today war fighting is about 16 times more energy-intensive then the Vietnam War era. Furthermore, oil intensity per war fighter rose 2.6 percent annually for the past 40 years and is anticipated to increase another 1.5 percent annually through 2017 due to greater mechanization, remote expeditionary conflict, rugged terrain, and irregular operations (Lovins). Energy became a vital fighting source that it is closely related to the troops’ survival and success in the battlefield. The paper concludes by arguing that it is fundamental to emphasis the issues presented and because of the energy crisis we are experiencing today there has never been a better time to pursue and consequently execute a concrete energy course of action applied to our military forces that can minimize our troops’ risks and strengthen our national security and global interests.Author's Origina
Improved Thermal Area Law and Quasilinear Time Algorithm for Quantum Gibbs States
No full textOne of the most fundamental problems in quantum many-body physics is the characterization of correlations among thermal states. Of particular relevance is the thermal area law, which justifies the tensor network approximations to thermal states with a bond dimension growing polynomially with the system size. In the regime of sufficiently low temperatures, which is particularly important for practical applications, the existing techniques do not yield optimal bounds. Here, we propose a new thermal area law that holds for generic many-body systems on lattices. We improve the temperature dependence from the original (β) to ̃ (β2/3), thereby suggesting diffusive propagation of entanglement by imaginary time evolution. This qualitatively differs from the real-time evolution which usually induces linear growth of entanglement. We also prove analogous bounds for the Rényi entanglement of purification and the entanglement of formation. Our analysis is based on a polynomial approximation to the exponential function which provides a relationship between the imaginary-time evolution and random walks. Moreover, for one-dimensional (1D) systems with n spins, we prove that the Gibbs state is well-approximated by a matrix product operator with a sublinear bond dimension of ẽ (βlog(n))√. This proof allows us to rigorously establish, for the first time, a quasi-linear time classical algorithm for constructing an MPS representation of 1D quantum Gibbs states at arbitrary temperatures of β=o(log(n)). Our new technical ingredient is a block decomposition of the Gibbs state, that bears resemblance to the decomposition of real-time evolution given by Haah et al., FOCS'18.Accepted Manuscrip
