This paper presents a computational method of analysis that draws from machine learning, library science, and literary studies to map the visual layouts of multi-ethnic newspapers from the late 19th and early 20th century United States. This work dep
arts from prior approaches to newspapers that focus on individual pieces of textual and visual content. Our method combines Chronicling Americas MARC data and the Newspaper Navigator machine learning dataset to identify the visual patterns of newspaper page layouts. By analyzing high-dimensional visual similarity, we aim to better understand how editors spoke and protested through the layout of their papers.
We consider Bayesian inference of sparse covariance matrices and propose a post-processed posterior. This method consists of two steps. In the first step, posterior samples are obtained from the conjugate inverse-Wishart posterior without considering
the sparse structural assumption. The posterior samples are transformed in the second step to satisfy the sparse structural assumption through the hard-thresholding function. This non-traditional Bayesian procedure is justified by showing that the post-processed posterior attains the optimal minimax rates. We also investigate the application of the post-processed posterior to the estimation of the global minimum variance portfolio. We show that the post-processed posterior for the global minimum variance portfolio also attains the optimal minimax rate under the sparse covariance assumption. The advantages of the post-processed posterior for the global minimum variance portfolio are demonstrated by a simulation study and a real data analysis with S&P 400 data.
We present a novel mechanism for Sommerfeld enhancement for dark matter interactions without the need for light mediators. Considering a model for two-component scalar dark matter with a triple coupling, we find that there appears an $u$-channel reso
nance in dark matter elastic scattering. From the sum of the corresponding ladder diagrams, we obtain a Bethe-Salpeter equation with a delay term and identify the Sommerfeld factor for two-component dark matter from the effective Yukawa potential for the first time. We discuss the implications of our results for enhancing dark matter self-scattering and annihilation.
Deep convolutional neural networks (CNNs) have shown state-of-the-art performances in various computer vision tasks. Advances on CNN architectures have focused mainly on designing convolutional blocks of the feature extractors, but less on the classi
fiers that exploit extracted features. In this work, we propose Split-and-Share Module (SSM),a classifier that splits a given feature into parts, which are partially shared by multiple sub-classifiers. Our intuition is that the more the features are shared, the more common they will become, and SSM can encourage such structural characteristics in the split features. SSM can be easily integrated into any architecture without bells and whistles. We have extensively validated the efficacy of SSM on ImageNet-1K classification task, andSSM has shown consistent and significant improvements over baseline architectures. In addition, we analyze the effect of SSM using the Grad-CAM visualization.
The extreme miniaturization of a cold-atom interferometer accelerometer requires the development of novel technologies and architectures for the interferometer subsystems. We describe several component technologies and a laser system architecture to
enable a path to such miniaturization. We developed a custom, compact titanium vacuum package containing a microfabricated grating chip for a tetrahedral grating magneto-optical trap (GMOT) using a single cooling beam. The vacuum package is integrated into the optomechanical design of a compact cold-atom sensor head with fixed optical components. In addition, a multichannel laser system driven by a single seed laser has been implemented with time-multiplexed frequency shifting using single sideband modulators, reducing the number of optical channels connected to the sensor head. This laser system architecture is compatible with a highly miniaturized photonic integrated circuit approach, and by demonstrating atom-interferometer operation with this laser system, we show feasibility for the integrated photonic approach. In the compact sensor head, sub-Doppler cooling in the GMOT produces 15 uK temperatures, which can operate at a 20 Hz data rate for the atom interferometer sequence. After validating atomic coherence with Ramsey interferometry, we demonstrate a light-pulse atom interferometer in a gravimeter configuration without vibration isolation for 10 Hz measurement cycle rate and T = 0 - 4.5 ms interrogation time, resulting in $Delta$g / g = 2.0e-6. All these efforts demonstrate progress towards deployable cold-atom inertial sensors under large amplitude motional dynamics.
In magnetic Weyl semimetals, fluctuations of the local magnetization may generate gauge fields that couple to the chiral charge of emergent Weyl fermions. Recent theoretical studies have proposed that the temporal and spatial-dependent magnetization
associated with propagating domain walls (DWs) generates pseudo electric and magnetic fields that drive novel phenomena such as a current of real charge. Here we report a key step in testing these predictions: characterizing the propagation of DWs in the Weyl semimetal Co3Sn2S2 using scanning magneto-optic Kerr microscopy. We observe an unexpected deep minimum in the temperature dependence of the DW mobility, $mu$, indicating a crossover between two regimes of propagation. The nonmonotonic $mu(T)$ is evidence of a phase transition in the topology of the DW well below the Curie temperature, in which the magnetization texture changes from continuous rotation (elliptical wall) to a linear wall whose unidirectional magnetization passes through zero at the wall center.
We introduce Causal Program Dependence Analysis (CPDA), a dynamic dependence analysis that applies causal inference to model the strength of program dependence relations in a continuous space. CPDA observes the association between program elements by constructing and executing modifi
To minimize enormous havoc from disasters, permanent environment monitoring is necessarily required. Thus we propose a novel energy management protocol for energy harvesting wireless sensor networks (EH-WSNs), named the adaptive sensor node managemen
t protocol (ASMP). The proposed protocol makes system components to systematically control their performance to conserve the energy. Through this protocol, sensor nodes autonomously activate an additional energy conservation algorithm. ASMP embeds three sampling algorithms. For the optimized environment sampling, we proposed the adaptive sampling algorithm for monitoring (ASA-m). ASA-m estimates the expected time period to occur meaningful change. The meaningful change refers to the distance between two target data for the monitoring QoS. Therefore, ASA-m merely gathers the data the system demands. The continuous adaptive sampling algorithm (CASA) solves the problem to be continuously decreasing energy despite of ASA-m. When the monitored environment shows a linear trend property, the sensor node in CASA rests a sampling process, and the server generates predicted data at the estimated time slot. For guaranteeing the self-sustainability, ASMP uses the recoverable adaptive sampling algorithm (RASA). RASA makes consumed energy smaller than harvested energy by utilizing the predicted data. RASA recharges the energy of the sensor node. Through this method, ASMP achieves both energy conservation and service quality.
We consider high-dimensional multivariate linear regression models, where the joint distribution of covariates and response variables is a multivariate normal distribution with a bandable covariance matrix. The main goal of this paper is to estimate
the regression coefficient matrix, which is a function of the bandable covariance matrix. Although the tapering estimator of covariance has the minimax optimal convergence rate for the class of bandable covariances, we show that it has a sub-optimal convergence rate for the regression coefficient; that is, a minimax estimator for the class of bandable covariances may not be a minimax estimator for its functionals. We propose the blockwise tapering estimator of the regression coefficient, which has the minimax optimal convergence rate for the regression coefficient under the bandable covariance assumption. We also propose a Bayesian procedure called the blockwise tapering post-processed posterior of the regression coefficient and show that the proposed Bayesian procedure has the minimax optimal convergence rate for the regression coefficient under the bandable covariance assumption. We show that the proposed methods outperform the existing methods via numerical studies.
We compute the expected sensitivity on measurements of optical depth to reionization for a ground-based experiment at Teide Observatory. We simulate polarized partial sky maps for the GroundBIRD experiment at the frequencies 145 and 220 GHz. We perfo
rm fits for the simulated maps with our pixel-based likelihood to extract the optical depth to reionization. The noise levels of polarization maps are estimated as 110 $mumathrm{K~arcmin}$ and 780 $ mumathrm{K~arcmin}$ for 145 and 220 GHz, respectively, by assuming a three-year observing campaign and sky coverages of 0.537 for 145 GHz and 0.462 for 220 GHz. Our sensitivities for the optical depth to reionization are found to be $sigma_tau$=0.030 with the simulated GroundBIRD maps, and $sigma_tau$=0.012 by combining with the simulated QUIJOTE maps at 11, 13, 17, 19, 30, and 40 GHz.
