La distribution des zéros des fonctions L

Selon la philosophie de Katz et Sarnak, la distribution des zéros des fonctions $L$ est prédite par le comportement des valeurs propres de matrices aléatoires. En particulier, le comportement des zéros près du point central révèle le type de symétrie de la famille de fonctions $L$. Une fois la symétrie identifiée, la philosophie de Katz et Sarnak conjecture que plusieurs statistiques associées aux zéros seront modélisées par les valeurs propres de matrices aléatoires du groupe correspondant. Ce mémoire étudiera la distribution des zéros près du point central de la famille des courbes elliptiques sur $\mathbb{Q}[i]$. Brumer a effectué ces calculs en 1992 sur la famille de courbes elliptiques sur $\mathbb{Q}$. Les nouvelles problématiques reliées à la généralisation de ses travaux vers un corps de nombres seront mises en évidence

Reconstructing the last interglacial at Summit, Greenland: Insights from GISP2

The Eemian (last interglacial, 130-115 ka) was likely the warmest of all interglacials of the last 800 ka, with summer Arctic temperatures 3-5 degrees C above present. Here, we present improved Eemian climate records from central Greenland, reconstructed from the base of the Greenland Ice Sheet Project 2 (GISP2) ice core. Our record comes from clean, stratigraphically disturbed, and isotopically warm ice from 2,750 to 3,040 m depth. The age of this ice is constrained by measuring CH_4 and delta O^18 of O_2, and comparing with the historical record of these properties from the North Greenland Ice Core Project (NGRIP) and North Greenland Eemian Ice Drilling (NEEM) ice cores. The d^18 O_ice, d^15N of N_2, and total air content for samples dating discontinuously from 128 to 115 ka indicate a warming of similar to 6 degrees C between 127-121 ka, and a similar elevation history between GISP2 and NEEM. The reconstructed climate and elevation histories are compared with an ensemble of coupled climate-ice-sheet model simulations of the Greenland ice sheet. Those most consistent with the reconstructed temperatures indicate that the Greenland ice sheet contributed 5.1 m (4.1-6.2 m, 95% credible interval) to global eustatic sea level toward the end of the Eemian. Greenland likely did not contribute to anomalously high sea levels at ~127 ka, or to a rapid jump in sea level at ~120 ka. However, several unexplained discrepancies remain between the inferred and simulated histories of temperature and accumulation rate at GISP2 and NEEM, as well as between the climatic reconstructions themselves.

Antidepressant effects of exercise: a role for the adiponectin-PGC-1α-kynurenine triad?

It is well-recognized that exercise improves mental health, e.g., by decreasing depressive behaviors, improving hippocampal-dependent learning and neurogenesis, and increasing dendritic plasticity. Yet how exercise influences the brain at the molecular level is not clearly understood. Yau et al recently reported that the antidepressant effects of physical exercise are mainly mediated by adiponectin, an adipocyte-secreted hormone ('adipocytokine') with neuroprotective effects at the central nervous system level (Yau et al., 2014). This article is protected by copyright. All rights reserved

Improved model of an electric calciner for carbon materials

Teknova have 2D steady-state models of the calciner but wish, in the long term, to have a 3D model that can also cover unsteady conditions, and can can model the loss of axisymmetry that someties occurs. Teknova also wish to understand the processes happening around the tip of the upper electrode, in particular the formation of a lip on it and the the shape of the empty region below it. The Study Group proposed potential models for the degree of graphitization, and for the granular flow. Also the Study Group considered the upper electrode in detail. The proposed model for the lip formation is by sublimation of carbon from the hottest parts of the furnace with redeposition in the region around the electrode, which may stick particles onto the electrode surface. In this model the region below the electrode would be a void, roughly a vertex-down conical cavity. The electric field near the lower rim of the electrode will then have a singularity and so the most intense heating of the charge will be around the rim. We conjecture that the reason why the lower electrode lasts so much longer than the upper is that it is not adjacent to a cavity like this, and therefore does not have a singularity in the field.

Scalar turbulent behavior in the roughness sublayer of an Amazonian forest.

An important current problem in micrometeorology is the characterization of turbulence in the roughness sublayer (RSL), where most of the measurements above tall forests are made. There, scalar turbulent fluctuations display significant departures from the predictions of Monin?Obukhov similarity theory (MOST). In this work, we analyze turbulence data of virtual temperature, carbon dioxide, and water vapor in the RSL above an Amazonian forest (with a canopy height of 40 m), measured at 39.4 and 81.6 m above the ground under unstable conditions. We found that dimensionless statistics related to the rate of dissipation of turbulence kinetic energy (TKE) and the scalar variance display significant departures from MOST as expected, whereas the vertical velocity variance follows MOST much more closely. Much better agreement between the dimensionless statistics with the Obukhov similarity variable, however, was found for the subset of measurements made at a low zenith angle Z, in the range 0°  <  |Z|  <  20°. We conjecture that this improvement is due to the relationship between sunlight incidence and the ?activation?deactivation? of scalar sinks and sources vertically distributed in the forest. Finally, we evaluated the relaxation coefficient of relaxed eddy accumulation: it is also affected by zenith angle, with considerable improvement in the range 0°  <  |Z|  <  20°, and its values fall within the range reported in the literature for the unstable surface layer. In general, our results indicate the possibility of better stability-derived flux estimates for low zenith angle ranges.

Scalar turbulent behavior in the roughness sublayer of an Amazonian forest.

An important current problem in micrometeorology is the characterization of turbulence in the roughness sublayer (RSL), where most of the measurements above tall forests are made. There, scalar turbulent fluctuations display significant departures from the predictions of Monin?Obukhov similarity theory (MOST). In this work, we analyze turbulence data of virtual temperature, carbon dioxide, and water vapor in the RSL above an Amazonian forest (with a canopy height of 40?m), measured at 39.4 and 81.6?m above the ground under unstable conditions. We found that dimensionless statistics related to the rate of dissipation of turbulence kinetic energy (TKE) and the scalar variance display significant departures from MOST as expected, whereas the vertical velocity variance follows MOST much more closely. Much better agreement between the dimensionless statistics with the Obukhov similarity variable, however, was found for the subset of measurements made at a low zenith angle Z, in the range 0°??conjecture that this improvement is due to the relationship between sunlight incidence and the ?activation?deactivation? of scalar sinks and sources vertically distributed in the forest. Finally, we evaluated the relaxation coefficient of relaxed eddy accumulation: it is also affected by zenith angle, with considerable improvement in the range 0°??derived flux estimates for low zenith angle ranges.

Essays in Gender Economics

This thesis consists of three papers on gender economics. Chapter 1 studies whether people dislike collaborating with someone who corrects them and whether the dislike is stronger when that person is a woman. Having a good relationship with colleagues is integral in group work, potentially leading to successful collaborations. However, there are occasions when people have to correct their colleagues. Using a quasi-laboratory experiment, I find that people, including those with high productivity, are less willing to collaborate with a person who has corrected them even if the correction improves group performance. In addition, I find suggestive evidence that men respond more negatively to women’s corrections, which is not driven by their beliefs about the difference in women’s and men’s abilities. These findings suggest that there is a behavioral bias in group work that distorts the optimal selection of talents and penalizes those who correct others’ mistakes, and the distortion may be stronger when women correct men. Chapter 2 studies the role of gender and cognitive skills on other peoples’ generosity. Using a novel experimental design where I exogenously vary gender and cognitive skills and sufficiently powered analysis, I find neither the two attributes nor their interactions affect other people’s generosity; if anything, people are more generous to women with high potential. Chapter 3 studies how increased legal tolerance toward domestic violence affects married women’s welfare using the domestic violence decriminalization bill introduced to the Russian national congress in 2016. Using difference-in-differences and flexibly controlling for macroeconomic shocks, I find that the bill decreased married women’s life satisfaction and increased depression, especially among those with a college degree and a highly qualified white-collar occupation supposed to be more sensitive to gender regressive atmosphere. Consistent with this conjecture, people became more tolerant toward general and domestic violence after the bill.

On Reasonable Space and Time Cost Models for the λ-Calculus

Slot and van Emde Boas Invariance Thesis states that a time (respectively, space) cost model is reasonable for a computational model C if there are mutual simulations between Turing machines and C such that the overhead is polynomial in time (respectively, linear in space). The rationale is that under the Invariance Thesis, complexity classes such as LOGSPACE, P, PSPACE, become robust, i.e. machine independent. In this dissertation, we want to find out if it possible to define a reasonable space cost model for the lambda-calculus, the paradigmatic model for functional programming languages. We start by considering an unusual evaluation mechanism for the lambda-calculus, based on Girard's Geometry of Interaction, that was conjectured to be the key ingredient to obtain a space reasonable cost model. By a fine complexity analysis of this schema, based on new variants of non-idempotent intersection types, we disprove this conjecture. Then, we change the target of our analysis. We consider a variant over Krivine's abstract machine, a standard evaluation mechanism for the call-by-name lambda-calculus, optimized for space complexity, and implemented without any pointer. A fine analysis of the execution of (a refined version of) the encoding of Turing machines into the lambda-calculus allows us to conclude that the space consumed by this machine is indeed a reasonable space cost model. In particular, for the first time we are able to measure also sub-linear space complexities. Moreover, we transfer this result to the call-by-value case. Finally, we provide also an intersection type system that characterizes compositionally this new reasonable space measure. This is done through a minimal, yet non trivial, modification of the original de Carvalho type system.

Transizioni di fase nell'allenamento delle Macchine di Boltzmann Ristrette

In this thesis, we dealt with Restricted Boltzmann Machines with binary priors as models of unsupervised learning, analyzing the role of the number of hidden neurons on the amount of examples needed for a successful training. We simulated a teacher-student scenario and calculated the efficiency of the machine under the assumption of replica symmetry to study the location of the critical threshold beyond which learning begins. Our results confirm the conjecture that, in the absence of correlation between the weights of the data-generating machine, the critical threshold does not depend on the number of hidden units (as long as it is finite) and thus on the complexity of the data. Instead, the presence of correlation significantly reduces the amount of examples needed for training. We have shown that this effect becomes more pronounced as the number of hidden units increases. The entire analysis is supported by numerical simulations that corroborate the results.