Large Neighbor Research Articles

Linear and Nonlinear Growth Determinants: The Case of Mongolia and its Connection to China

We investigate growth determinants for Mongolia as a small emerging economy considering China as its large neighbor. Our causality analysis from January 1992 to August 2017 reveals significant linear and nonlinear relationships in growth explanation. China’s GDP and coal prices, together with some of their linear and nonlinear lagged components, predict Mongolia’s GDP, where a one percent increase in China’s GDP relates to an increase in Mongolia of 1.5 percent. Current exchange rates and the nonlinear components of lagged levels of consumer prices also explain growth. Our results underline the role of macroeconomic drivers of growth in emerging economies.

Behavior recognition based on track space-time characteristics

Under the influence of novel corona virus pneumonia epidemic prevention and control, higher requirements for behavior recognition in complex environment are put forward. The accuracy of traditional methods for sports training is not high, so a method is needed to improve the local action recognition to assist sports training. In the process of behavior recognition, if only the track is regarded as an independent individual, the information of its neighbor will be ignored. Therefore, we use KNN algorithm to get the nearest neighbor trajectory. In order to calculate the rich neighborhood information around the track, this paper calculates the complex relationship between the center track and the neighborhood track from four different angles, including absolute motion, relative motion, distance relationship and direction relationship. Then, from the four different perspectives of variance, discrete coefficient, skewness and kurtosis, this paper proposes a large interval nearest neighbor coding method. This method makes the four eigenvalues complement each other and improves the ability of describing complex and changeable behaviors. The experimental results show that the coding method proposed in this paper can be used for behavior recognition according to different transformation matrix.

The Effects of Health Spending on the Propagation of Infectious Diseases

Unraveling the relation of health spending and outcomes between bordering jurisdictions is a first-order concern to better coordinate public health provision. In this paper, we explore the rules governing the main federal grant to Brazilian municipalities, which varies discontinuously near population thresholds, to identify the effects of a subsequent exogenous increase in local health spending on bordering towns' health inputs and outcomes. We provide evidence that 1% increase in health spending reduces hospitalization for respiratory infectious diseases by 1 to 3%. Second, we find that both the smallest and largest neighbors benefit by reducing infant and elderly mortality and hospitalization for infectious diseases. Third, the only input that significantly increases is the number of physicians, corroborating that primary care by medical professionals helps to detain the proliferation an the worsening of infectious diseases. Finally, fiscal reactions are asymmetric: small jurisdictions reduce health spending while large neighbors spend more, mainly hiring more physicians.

The power of vision, leadership and consensus: Downtown Des Moines transformation 1980–2019

By the 1980s, downtown Des Moines had developed as a successful office centre, with major insurance companies headquartered there. Nevertheless, it lacked many of the attributes and qualitative elements that create a dynamic social and cultural place that would attract the talent companies needed, or have the makings of a great place dear to the community. There was little downtown housing, no pedestrian activity, few retail outlets and quality restaurants were few. Supportive public transport, walkability and bikeability had not been successfully addressed. In short, downtown vitality was sorely missing; however, local civic and business leadership engaged in extensive and intensive planning and strategic initiatives. The process lasted a number of years, guided by a desire to bring together all the elements of vibrancy to make Des Moines more competitive with its larger neighbours. This paper outlines the planning and implementation that literally transformed a quiet but uncompelling downtown into one of the most celebrated and award-winning downtowns in America. The engagement and support of the broader community was critical in achieving this vision. The assistance provided by local talent and outside consultants was essential in guiding the revitalisation process. The result has been a multi-billion-dollar transformation that occurred over a period of three decades.

Learning a Distance Metric by Balancing KL-Divergence for Imbalanced Datasets

In many real-world domains, datasets with imbalanced class distributions occur frequently, which may confuse various machine learning tasks. Among all these tasks, learning classifiers from imbalanced datasets is an important topic. To perform this task well, it is crucial to train a distance metric which can accurately measure similarities between samples from imbalanced datasets. Unfortunately, existing distance metric methods, such as large margin nearest neighbor, information-theoretic metric learning, etc., care more about distances between samples and fail to take imbalanced class distributions into consideration. Traditional distance metrics have natural tendencies to favor the majority classes, which can more easily satisfy their objective function. Those important minority classes are always neglected during the construction process of distance metrics, which severely affects the decision system of most classifiers. Therefore, how to learn an appropriate distance metric which can deal with imbalanced datasets is of vital importance, but challenging. In order to solve this problem, this paper proposes a novel distance metric learning method named distance metric by balancing KL-divergence (DMBK). DMBK defines normalized divergences using KL-divergence to describe distinctions between different classes. Then it combines geometric mean with normalized divergences and separates samples from different classes simultaneously. This procedure separates all classes in a balanced way and avoids inaccurate similarities incurred by imbalanced class distributions. Various experiments on imbalanced datasets have verified the excellent performance of our novel method.

Social selection acts on behavior and body mass but does not contribute to the total selection differential in eastern chipmunks.

Through social interactions, phenotypes of conspecifics can affect an individual's fitness, resulting in social selection. Social selection is assumed to represent a strong and dynamic evolutionary force that can act with or in opposition to natural selection. Few studies, however, have estimated social selection and its contribution to total selection in the wild. We estimated natural and social selection gradients on exploration, docility, and body mass, and their contribution to selection differentials, in a wild eastern chipmunk population (Tamias striatus). We applied trait-based multiple regression models derived from classical phenotypic selection analyses, which allowed us to include several social partners (i.e., neighbors). We detected social selection gradients on female docility and male body mass, indicating that female with docile neighbors and males with large neighbors had lower fitness. In both sexes, social selection gradients varied with the season. However, we found no phenotypic assortment or disassortment for the studied traits. Social selection gradients, therefore, did not contribute to total selection differentials, and natural selection alone could drive phenotypic changes. Evaluating the factors that drive the evolution of the covariance between interacting phenotypes is necessary to understand the role of social selection as an evolutionary force.

Influence of Effective Water-to-Cement Ratios on Internal Damage and Salt Scaling of Concrete with Superabsorbent Polymer.

Superabsorbent polymer (SAP) is attracting attention as a water-entraining admixture that reduces shrinkage or heals cracks in concrete. Cross-linked sodium polyacrylate SAPs, which are the most widely produced SAPs in the global market, are applicable as concrete admixtures. However, there have been contradictory results on the freeze–thaw resistance of concrete with SAPs. This study aims to clarify these results considering the water absorption behavior of SAPs in hardened concrete when effective water-to-cement ratios are different. Firstly, the absorbencies of one kind of cross-linked sodium polyacrylate SAP (SAP_SP) in pore solution and fresh mortar were measured by a tea bag test and flow test, respectively. Pore size distribution, capillary water absorption, and deformation during freeze–thaw cycles were analyzed for mortar samples with varying SAP_SP dosages. In the main tests, concrete samples with three different SAP_SPs/cement ratios (0.1%, 0.2%, and 0.3%) and a reference sample were prepared, and internal damage and salt scaling were measured under freeze–thaw cycles. Because SAP_SP absorbs water in fresh mixtures, additional water was added to the mixture considering the water absorbency of the SAP_SP. It was found that the used SAP_SPs prematurely release their stored water so the effective water-to-cement ratio was increased when a larger amount of SAP_SP was used. The higher effective water-to-cement ratio caused more internal damage and salt scaling due to the weaker cementitious matrix. In addition, mortar samples with a high SAP_SP content show a larger absorption of capillary water than the reference sample. The result can be interpreted by an observation that SAP_SP in air voids absorbs water and expands to relatively large capillary pores or neighbor air voids during the capillary water absorption process.

Link prediction based on the tie connection strength of common neighbor

Traditional link prediction indices focus on the degree of the common neighbor and consider that the common neighbor with large degree contributes less to the similarity of two unconnected endpoints. Therefore, some of the local information-based methods only restrain the common neighbor with large degree for avoiding the influence dissipation. We find, however, if the large degree common neighbor connects with two unconnected endpoints through multiple paths simultaneously, these paths actually serve as transmission influences instead of dissipation. We regard these paths as the tie connection strength (TCS) of the common neighbor, and larger TCS can promote two unconnected endpoints to link with each other. Meanwhile, we notice that the similarity of node-pairs also relates to the network topology structure. Thus, in order to study the influences of TCS and the network structure on similarity, we introduce a free parameter and propose a novel link prediction method based on the TCS of the common neighbor. The experiment results on 12 real networks suggest that the proposed TCS index can improve the accuracy of link prediction.

Energy Diplomacy In Azerbaijan’s Foreign Policy

In the 21st century, the relations between states are being shaped by factors deriving from the national interest and the national security agenda. Many countries have shifted their foreign policies towards an economization strategy, i.e. assigning a particular importance to the economic priorities and instruments in foreign policy, as in the case of the small state of Azerbaijan, bordered by large neighbors and influenced by yet larger powers. Before Azerbaijan’s independence, Regional powers such as Russia, Turkey and Iran were active and after the independence, the US and EU, added to the competition over the rich oil and natural gas resources of the Caspian region. All of these countries created an external pressure and influence in Azerbaijan, turning the Caucasus region into a new playing field of the Great Power Game. Thus, the geopolitical ambiguity combined with oil and gas resources and further with geographical imperatives will shape and pose challenge to the country’s foreign policy.

A LightGBM-Based EEG Analysis Method for Driver Mental States Classification.

Fatigue driving can easily lead to road traffic accidents and bring great harm to individuals and families. Recently, electroencephalography- (EEG-) based physiological and brain activities for fatigue detection have been increasingly investigated. However, how to find an effective method or model to timely and efficiently detect the mental states of drivers still remains a challenge. In this paper, we combine common spatial pattern (CSP) and propose a light-weighted classifier, LightFD, which is based on gradient boosting framework for EEG mental states identification. The comparable results with traditional classifiers, such as support vector machine (SVM), convolutional neural network (CNN), gated recurrent unit (GRU), and large margin nearest neighbor (LMNN), show that the proposed model could achieve better classification performance, as well as the decision efficiency. Furthermore, we also test and validate that LightFD has better transfer learning performance in EEG classification of driver mental states. In summary, our proposed LightFD classifier has better performance in real-time EEG mental state prediction, and it is expected to have broad application prospects in practical brain-computer interaction (BCI).

Cell-Size Pleomorphism Drives Aberrant Clone Dispersal in Proliferating Epithelia.

As epithelial tissues develop, groups of cells related by descent tend to associate in clonal populations rather than dispersing within the cell layer. While this is frequently assumed to be a result of differential adhesion, precise mechanisms controlling clonal cohesiveness remain unknown. Here we employ computational simulations to modulate epithelial cell size in silico and show that junctions between small cells frequently collapse, resulting in clone-cell dispersal among larger neighbors. Consistent with similar dynamics invivo, we further demonstrate that mosaic disruption of Drosophila Tor generates small cells and results in aberrant clone dispersal in developing wing disc epithelia. We propose a geometric basis for this phenomenon, supported in part by the observation that soap-foam cells exhibit similar size-dependent junctional rearrangements. Combined, these results establish a link between cell-size pleomorphism and the control of epithelial cell packing, with potential implications for understanding tumor cell dispersal in human disease.

Mexico and Uruguay inbound tourism demand


 
 
 
 The main objective of this paper is to estimate the tourism demand for Mexico and Uruguay, two very different countries, but for both of which tourism is an important activity, and mainly originating from a large neighbor. We try to analyze whether the determinants of tourism demand differ depending on the size of the country, or if being a neighboring country is the main determinant. So, we analyze the relationship between the number of American tourists visiting Mexico and Argentinian tourists visiting Uruguay, and the inbound tourists’ income and the bilateral real exchange rate (RER) between the visiting country and the hosting country, following the Johansen’s methodology. We found one cointegration relationship for each country, where the income-elasticity was greater than 2 for American tourists visiting Mexico, and nearly 3 for Argentinian tourists visiting Uruguay. Bilateral RERs were also significant in both models. Moreover, forecasts show the impacts of institutional changes on the tourism sector. The impact of arrival of President Macri to power was positive for Argentinian tourists visiting Uruguay, but President Trump's arrival in the US was negative for American tourists visiting Mexico.
 
 
 

Large Margin Nearest Neighbor Classification With Privileged Information for Biometric Applications

In this paper, a new metric learning algorithm is proposed to improve face verification and person re-identification in RGB images by learning from RGB and Depth (RGB-D) training images. We address this problem by formulating it as a Learning Using Privileged Information problem, in which the additional depth images associated with the RGB training images are not available for the testing process. Based on the large margin nearest neighbor (LMNN) classification framework, we propose an effective metric learning method called large margin nearest neighbor classification with privileged information (LMNN+) by incorporating depth information to improve decision function learning in the training process. Specifically, two distance metrics based on visual features as well as depth features are jointly learned by minimizing the triplet loss in which the within-class difference is minimized, while the between-class difference is maximized. The distances in the depth feature space can be utilized to guide the training process in the visual feature space. In addition, we propose an efficient optimization method which can handle billions of constraints in the optimization problem of LMNN+. The comprehensive experiments on the EUROCOM data set, the CurtinFaces data set as well as the BIWI RGBD-ID data set demonstrate the effectiveness of our algorithm for face verification and person re-identification by leveraging privileged information.

Real-Time Detection of Fall From Bed Using a Single Depth Camera

Toward the medical and living healthcare for the elderly and patients, fall from bed is a critical accident that may lead to serious injuries. To alleviate this, an essential problem is to detect this event in time for earning the rescue time. Although some efforts that resort to the wearable devices and smart healthcare room have already been paid to address this problem, the performance is still not satisfactory enough for the practical applications. In this paper, a novel fall from a bed detection method is proposed. In particular, the depth camera is used as the visual sensor due to its insensitivity to illumination variation and capacity of privacy protection. To characterize the human activity well, an effective human upper body detection approach able to extract human head and upper body center is proposed using random forest. Compared with the existing widely used human body parsing methods (e.g., Microsoft Kinect SDK or OpenNI SDK), our proposition can still work reliably when human–bed interaction happens. According to the motion information of human upper body, the fall from bed detection task is formulated as a two-class classification problem. Then, it is solved using the large margin nearest neighbor classification approach. Our method can meet the real-time running requirement with the normal computer. In experiments, we construct a fall from bed detection data set that contains the samples from 42 volunteers (26 males and 16 females) for test. The experimental results demonstrate the effectiveness and efficiency of our proposition. Note to Practitioners—This paper aims to develop the vision-based surveillance system to automatically detect fall from bed in real time under the unconstrained conditions. This cannot be well handled by using wearable devices. Our proposition can be used to construct a smart healthcare room. It is worth noting that the proposed system is insensitive to illumination variation and capable of protecting privacy due to the use of depth camera. The proposed human body extraction method is able to address the human–bed interaction well. An effective motion pattern categorization approach is applied to ensure the good separation between fall from bed and the other activities. According to the extensive experiments under the laboratory and sickroom environments, our fall from bed detection method generally can achieve acceptable result with high efficiency. Nevertheless, it is still somewhat sensitive to noise within the depth frames. In addition, when serious human–quilt interaction happens, the performance of human body extraction procedure is not satisfactory enough.

Bootstrapping search margin-based nearest neighbor method for qualitative spectroscopic analysis.

Qualitative spectroscopic analysis depends in one way or another on comparing spectra of the specimens to be identified with spectra of "known" or "standard" samples. The k-nearest neighbor (k-NN) method is one of the oldest and simplest techniques for performing such comparisons. In this study, we present a new k-NN algorithm for qualitative spectroscopic analysis, which we refer to as the bootstrapping search margin-based nearest neighbor (BSMNN) method. This method consists of two phases. In the first phase, we attempt to find a feature space in which samples with different labels produce large margins, such that the classification has high confidence, by maximizing a margin-based objective function using a weighted bootstrap sampling search strategy. In the subsequent classification phase, we compute local Euclidean distances between different samples under the feature space. A new instance x is classified by the label of its nearest neighbor. Six widely used vibrational spectroscopic data sets were used to validate the performance of the BSMNN method. The results showed that, despite its simplicity, BSMNN yields better results compared with commonly used k-NN algorithms including Relief, neighborhood components analysis, neighborhood component feature selection, and large margin nearest neighbor. Furthermore, BSMNN can be used to identify important spectral regions. It is worth mentioning that the margin-based objective function used in BSMNN is proposed for the first time for measuring the quality of features. Although in this study the margin-based objective function is focused on k-NN classification, it also can be used for other distance-based classifiers, such as soft independent modeling of class analogies and least squares support vector machine.

Revisiting the Kitaev material candidacy of Ir4+ double perovskite iridates

Quantum magnets with significant bond-directional Ising interactions, so-called Kitaev materials, have attracted tremendous attention recently in the search for exotic spin liquid states. Here we present a comprehensive set of measurements that enables us to investigate the crystal structures, Ir$^{4+}$ single ion properties, and magnetic ground states of the double perovskite iridates La$_2B$IrO$_6$ ($B$ $=$ Mg, Zn) and $A_2$CeIrO$_6$ ($A$ $=$ Ba, Sr) with a large nearest neighbor distance $>$ 5 Angstroms between Ir$^{4+}$ ions. Our neutron powder diffraction data on Ba$_2$CeIrO$_6$ can be refined in the cubic space group Fm$\bar{3}$m, while the other three systems are characterized by weak monoclinic structural distortions. Despite the variance in the non-cubic crystal field experienced by the Ir$^{4+}$ ions in these materials, X-ray absorption spectroscopy and resonant inelastic x-ray scattering are consistent with $J_{\rm eff}$ $=$ 1/2 moments in all cases. Furthermore, neutron scattering and resonant magnetic x-ray scattering show that these systems host A-type antiferromagnetic order. These electronic and magnetic ground states are consistent with expectations for face-centered-cubic magnets with significant antiferromagnetic Kitaev exchange, which indicates that spacing magnetic ions far apart may be a promising design principle for uncovering additional Kitaev materials.

Neighborhood competition mediates crown development of Picea sitchensis in Olympic rainforests: Implications for restoration management

Large trees are critically important for structuring ecosystems and providing habitat, and trees with complex crowns provide more of these services than comparably sized trees with simple crowns. Forest managers are increasingly emulating old-growth structure by retaining various densities of aggregated and dispersed trees. This study explores how individual within-crown complexity develops in a wide range of tree densities and interprets them in terms of management goals. Stratified LiDAR data were used to select 36 Picea sitchensis trees 55 to 91 m tall and from the complete range of tree densities in Olympic rainforests. These were climbed, three-dimensionally mapped—including the trunk and every appendage—to quantify tree allometry and habitat-related structures within their crowns. Sampled trees were then aged with crossdated increment cores to construct a chronosequence of crown development. The sample ranged from 70 to 250 cm diameter and 100 to 400 years old. Circular 30 m radius plots were installed around each tree to quantify neighborhood density. An additional 19 trees ranging from 10 to 94 m tall were added to the sample (N = 55) to describe baseline crown development and then compared to the 36-tree sample in which we modeled how this baseline is altered by neighborhood competition. After ∼200 yr, crown mass increases faster than trunk mass due to accumulation and growth of large appendages. The largest neighbors within 25 m compete with Picea, reducing trunk and appendage diameters as well as crown volume by roughly 20% for each 25% increase in density for a given tree age. Reduced growth and higher mortality of lower-crown appendages in dense forests delays development of large (>15 cm diameter) appendages for over 100 yr. Abundance of structures associated with reiterated trunks is unrelated to neighborhood density but increases with crown damage. These findings contribute to an empirically-based conceptual framework for crown development in Picea and interpretation of management guidelines. Picea is then compared to Pseudotsuga menziesii—another well-studied tall conifer of Olympic rainforests. Picea accumulates ecologically important crown structures much faster than Pseudotsuga yet has roughly half the longevity, indicating that these co-occurring species have complementary functions. Because of its shade tolerance and extremely rapid growth in rainforests, Picea can be used to accelerate development of complex crown structures in forests typically planted with Pseudotsuga after logging.

‘In the interest of peace and quiet in Europe’: The military and strategic role of Dutch neutrality, 1890-1940

The Netherlands is positioned amidst three major powers and controls the mouths of three main European rivers. Until the First World War, its choice for armed neutrality (1840-1940) seemed to be the most fitting answer to its security problem. After 1918, the Netherlands had difficulties adjusting to modern war, having decreased its defence budget substantially, and lacked a coherent political-military answer to the interwar strategic and operational challenges. Old notions of the Netherlands as a vital element of regional peace and as a country that could influence the behaviour of its large neighbours no longer fitted reality. Neutrality ceased to provide security to the country, thereby also endangering the stability in Western Europe to which the Dutch so wholeheartedly aspired.

Heterogeneous Activation, Local Structure, and Softness in Supercooled Colloidal Liquids.

We experimentally characterize heterogeneous nonexponential relaxation in bidisperse supercooled colloidal liquids utilizing a recent concept called "softness" [Phys. Rev. Lett. 114, 108001 (2015)PRLTAO0031-900710.1103/PhysRevLett.114.108001]. Particle trajectory and structure data enable classification of particles into subgroups with different local environments and propensities to hop. We determine residence times t_{R} between particle hops and show that t_{R} derived from particles in the same softness subgroup are exponentially distributed. Using the mean residence time t[over ¯]_{R} for each softness subgroup, and a Kramers' reaction rate model, we estimate the activation energy barriers E_{b} for particle hops, and show that both t[over ¯]_{R} and E_{b} are monotonic functions of softness. Finally, we derive information about the combinations of large and small particle neighbors that determine particle softness, and we explicitly show that multiple exponential relaxation channels in the supercooled liquid give rise to its nonexponential behavior.

Efficient Linear Feature Extraction Based on Large Margin Nearest Neighbor

Efficient Linear Feature Extraction Based on Large Margin Nearest Neighbor