Dependency Relations Research Articles

DUCD: Deep Unfolding Convolutional-Dictionary network for pansharpening remote sensing image

The goal of pansharpening methods is to complement the spectral and spatial information contained in Multi-spectral (MS) and panchromatic (PAN) images to obtain the desired High-resolution multispectral (HRMS) image. The existing majority of pansharpening methods either extract feature information separately from the MS image and PAN image, or extract feature information after concatenating the MS image and PAN image. However, the entire extraction process lacks the utilization of complementary information and tends to generate redundant information, thereby leading to the loss of certain important information during the extraction process, which in turn affects the overall performance. In order to better utilize the complementary information between the MS image and PAN image and enhance the interpretability of the network, we propose the Deep Unfolding Convolutional-Dictionary Network (DUCD) for pansharpening in this paper. This network fully integrates complementary information between the MS image and PAN image to generate the final fused image. The entire network structure consists of two parts: The encoder and the decoder. In the encoder part of the network, we clarify the common and unique feature information between MS and PAN images by constructing an observation model. Simultaneously, we use the approximate gradient algorithm to continuously optimize the model and iteratively unfold it into a deep network structure. In the decoder part of the network, we concatenate the obtained common and specific information from MS and PAN images and pass them through convolutional and activation layers. Subsequently, they are input into the introduced Frequency Domain-based Transformer (FDT) module and an information-lossless inversible neural network(INN). This provides a more efficient method for establishing long-range dependency relationships between feature extraction and feature fusion. To demonstrate the effectiveness of our proposed method, we conduct extensive experiments on three benchmark datasets QB, GF2 and WV3. Experimental results show that our method outperforms the current SOTA Pansharpening methods in terms of performance.

Slaves Holding Slaves: Mükâtebe Contracts, Velâ and a Probate Inventory in the Seventeenth-Century Crimean Khanate

Abstract This article aims to reveal some aspects of relationships between former slaves and their ex-owners in light of seventeenth-century Crimean qadi court records. It elaborates on a number of terms that indicate the social and legal status of slaves in the Crimean Khanate and analyzes a former slave probate inventory. In addition, the paper also examines the phenomenon of the mükâtebe contract in the Crimean Khanate context which allowed slaves to hold slaves; it thus seeks to provide new perspectives on the exercising of the mükâtebe. The paper considers how social fluidity affected the lives of slaves, explores the question of whether manumitted slaves actually attained the same status as the freeborn, and, finally, traces the ways in which dependency relationships evolved between freed former slaves and the slaves held by these latter.

Spatiotemporal correlation based self-adaptive pose estimation in complex scenes

In the current digital era, large-scale Artificial Intelligence (AI) models have fundamentally changed the landscape of the AI field. These models excel at extracting complex patterns and information from massive datasets, making them indispensable in a wide range of domains. With the advancement of large-scale AI models, pose estimation has achieved improved performance in managing large data sets and complex computations. However, most existing pose estimation methods suffer from the problem of inaccurate pose graph extraction in complex scenes. Moreover, the existing methods all focus on the message, ignoring the fact that it may not be optimal for the graph structure. This fixed graph structure is difficult to reflect the real dependency relationship between joint points. This study proposes a Character Redetection Model (CRM) to dynamically adjust the foreground extraction strategies to solve these problems. The CRM locates the target position based on the rich spatiotemporal information between the adjacent frames, effectively solving the problem of inaccurate foreground acquisition caused by occlusion and blur in complex scenes. In addition, an Uncertainty Graph Structure (UGS) is designed, which adaptively learns the topological structure of graphs with different Graph Convolutional Network (GCN) layers and skeleton samples in an end-to-end manner and mines the hidden space dependency between data to accurately model the different poses of the characters. The experimental results show that the proposed method achieves a superior performance compared to that of the State-Of-The-Art (SOTA) method, with a Percentage of Correct Keypoints (PCK) value of 94.6% at the Max Planck Institute for Informatics (MPII), along with the mean Average Precision (mAP) values of 83.3% and 82.1% on the PoseTrack2017 and classroom pose datasets, respectively.

Gender Issues in Digital Financial Literacy and Financial Behavior among Millennials

AbstractThe study examines the effect of digital financial literacy on financial behavior which is proxied by spending and saving behavior. In particular, this study investigates the role of gender differences in the dependency relationship between digital financial literacy on financial behavior. The population was undergraduate students of the Faculty of Economics, Sriwijaya University, Indonesia. We employed the snowball sampling method and obtained 136 respondents as samples. Ordinary Least Square and Pairwise Comparisons were used to test the hypotheses. The results show that digital financial literacy has a positive and significant effect on spending behavior and saving behavior, respectively. Moreover, there is a difference between men and women in the effect of digital financial literacy on spending behavior. Nevertheless, it is found that there is no difference between men and women in the effect of digital financial literacy on saving behavior. Keywords:Digital Financial Literacy, Spending Behavior, Saving Behavior, Financial Behavior, Gender, Millennials

A Multi-Stage Feature Aggregation and Structure Awareness Network for Concrete Bridge Crack Detection.

One of the most significant problems affecting a concrete bridge's safety is cracks. However, detecting concrete bridge cracks is still challenging due to their slender nature, low contrast, and background interference. The existing convolutional methods with square kernels struggle to capture crack features effectively, fail to perceive the long-range dependencies between crack regions, and have weak suppression ability for background noises, leading to low detection precision of bridge cracks. To address this problem, a multi-stage feature aggregation and structure awareness network (MFSA-Net) for pixel-level concrete bridge crack detection is proposed in this paper. Specifically, in the coding stage, a structure-aware convolution block is proposed by combining square convolution with strip convolution to perceive the linear structure of concrete bridge cracks. Square convolution is used to capture detailed local information. In contrast, strip convolution is employed to interact with the local features to establish the long-range dependence relationship between discrete crack regions. Unlike the self-attention mechanism, strip convolution also suppresses background interference near crack regions. Meanwhile, the feature attention fusion block is presented for fusing features from the encoder and decoder at the same stage, which can sharpen the edges of concrete bridge cracks. In order to fully utilize the shallow detail features and deep semantic features, the features from different stages are aggregated to obtain fine-grained segmentation results. The proposed MFSA-Net was trained and evaluated on the publicly available concrete bridge crack dataset and achieved average results of 73.74%, 77.04%, 75.30%, and 60.48% for precision, recall, F1 score, and IoU, respectively, on three typical sub-datasets, thus showing optimal performance in comparison with other existing methods. MFSA-Net also gained optimal performance on two publicly available concrete pavement crack datasets, thereby indicating its adaptability to crack detection across diverse scenarios.

WFSS: weighted fusion of spectral transformer and spatial self-attention for robust hyperspectral image classification against adversarial attacks

The emergence of adversarial examples poses a significant challenge to hyperspectral image (HSI) classification, as they can attack deep neural network-based models. Recent adversarial defense research tends to establish global connections of spatial pixels to resist adversarial attacks. However, it cannot yield satisfactory results when only spatial pixel information is used. Starting from the premise that the spectral band is equally important for HSI classification, this paper explores the impact of spectral information on model robustness. We aim to discover potential relationships between different spectral bands and establish global connections to resist adversarial attacks. We design a spectral transformer based on the transformer structure to model long-distance dependency relationships among spectral bands. Additionally, we use a self-attention mechanism in the spatial domain to develop global relationships among spatial pixels. Based on the above framework, we further explore the influence of both spectral and spatial domains on the robustness of the model against adversarial attacks. Specifically, a weighted fusion of spectral transformer and spatial self-attention (WFSS) is designed to achieve the multi-scale fusion of spectral and spatial connections, which further improves the model’s robustness. Comprehensive experiments on three benchmarks show that the WFSS framework has superior defensive capabilities compared to state-of-the-art HSI classification methods.

Multi-Task Learning with Sequential Dependence Toward Industrial Applications: A Systematic Formulation

Multi-task learning (MTL) is widely used in the online recommendation and financial services for multi-step conversion estimation, but current works often overlook the sequential dependence among tasks. In particular, sequential dependence multi-task learning (SDMTL) faces challenges in dealing with complex task correlations and extracting valuable information in real-world scenarios, leading to negative transfer and a deterioration in the performance. Herein, a systematic learning paradigm of the SDMTL problem is established for the first time, which applies to more general multi-step conversion scenarios with longer conversion paths or various task dependence relationships. Meanwhile, an SDMTL architecture, named Task-Aware Feature Extraction (TAFE), is designed to enable the dynamic task representation learning from a sample-wise view. TAFE selectively reconstructs the implicit shared information corresponding to each sample case and performs the explicit task-specific extraction under dependence constraints, which can avoid the negative transfer, resulting in more effective information sharing and joint representation learning. Extensive experiment results demonstrate the effectiveness and applicability of the proposed theoretical and implementation frameworks. Furthermore, the online evaluations at MYbank showed that TAFE had an average increase of 9.22% and 3.76% in various scenarios on the post-view click-through & conversion rate (CTCVR) estimation task. Currently, TAFE is deployed in an online platform to provide various traffic services.

Physical Activity Is Associated with the Incidence of Depression in United States Adults from the NHANES 2013-18: A Cross-Sectional Study.

The number of depression sufferers is rising globally. In the United States, 8% of adults over 20 years of age suffer from it, making it the most prevalent mental disorder in the country. Some lifestyle habits have been shown to favor or prevent the onset of depression; for instance, physical inactivity is associated with an increased likelihood of suffering depression, whilst multiple benefits have been attributed to performing physical activity (PA). This study aims to test whether there is a dependence between the prevalence of depression and PA, age, gender and educational level. The secondary objective was to identify the differentiating variables for depression and non-depression. This cross-sectional study is based on data from the NHANES 2013-2014, 2015-2016 and 2017-2018 editions. Some of the items in this survey were taken from preexisting questionnaires: the Patient Health Questionnaire-9 for depression screening and the Global Physical Activity Questionnaire (GPAQ) for the PA groups. The final sample was formed of 15,574 United States residents over 18 years old. After testing the data normality (p < 0.001), a descriptive analysis and the non-parametric chi-square test was conducted, as well as discriminant analysis. The results showed that there was an association between depression prevalence and PA (p < 0.001) in the general population and for both genders. Inactive participants had the highest prevalence of major depression and other depressive disorders. The discriminant analysis identified PA group (0.527), education level (0.761) and gender (-0.505) as significant variables that differentiate between participants with and without depression. The results of this research confirmed that a dependency relationship between PA group according to the GPAQ and depression prevalence according to the PHQ-9 existed in the United States adult population, and that PA group is a relevant variable to differentiate between depression sufferers and non-sufferers.

Investigation on fatigue life of push-the-bit rotary steerable bottom hole assembly with random bit-rock interaction

Rotary steerable systems are critical for challenging boreholes in deep-sea oil and gas drilling, where fatigue failure is the normal form of damage. A procedure with deterministic excitation and random excitation is presented in this paper to reliably predict the fatigue life of the bottom hole assembly. The initial defects, the Coulomb friction model, frictional contact between the drill bit and rock, and the stochastic rock-cutting process are included in the model. Monte Carlo approach is used to obtain the Statistical characteristics of the stochastic model. Based on the program mentioned above, influences of some parameters, such as rotational speed, weight on bit, steering force, and random excitation on torsional vibration characteristics, are investigated. Moreover, the rain-flow counting method is applied to reveal the dependency relationship between fatigue life and torsional vibration. It is found that random excitation can exacerbate the torsional vibration phenomenon and even significantly reduce the fatigue life. Finally, compared with the deterministic system, the sensitivity index coefficient of the fatigue life in the random excitation model is about 0.8 % - 2.2 % higher. The research results can provide a theoretical basis for reducing tool failure in on-site deep-sea oil and gas drilling.

Multi-Scale Non-Local Spatio-Temporal Information Fusion Networks for Multi-Step Traffic Flow Forecasting

Traffic flow prediction is a crucial research area in traffic management. Accurately predicting traffic flow in each area of the city over the long term can enable city managers to make informed decisions regarding the allocation of urban transportation resources in the future. The existing traffic flow prediction models either give insufficient attention to the interactions of long-lasting spatio-temporal regions or extract spatio-temporal features in a single scale, which ignores the identification of traffic flow patterns at various scales. In this paper, we present a multi-scale spatio-temporal information fusion model using non-local networks, which fuses traffic flow pattern features at multiple scales in space and time, complemented by non-local networks to construct the global direct dependence relationship between local areas and the entire region of the city in space and time in the past. The proposed model is evaluated through experiments and is shown to outperform existing benchmark models in terms of prediction performance.

APPLICATION OF DELTA GAMMA (THETA) NORMAL APPROXIMATION IN RISK MEASUREMENT OF AAPL'S AND GOLD'S OPTION

The option value has a nonlinear dependence relationship on risk factors existing in the capital market. Therefore, this paper considered utilizing Delta Gamma (Theta) Normal Approximation (DGTNA) as a nonlinear approach to determine the change of profit/loss of a European call option to assess the option risk. The method uses the second order of Taylor Polynomial around the stock price underlying the option to approximate the option profit/loss, which is crucial to construct the VaR based on DGTNA. VaR based on DGTNA also considered three Greeks, namely Delta, Gamma, and Theta, known as sensitivity measures in option. This research applied VaR based on DGTN approximation to analyze the European call option of Apple Inc (AAPL) and Barrick Gold Corporation (GOLD) for several strike prices. The performance of DGTN VaR analyzed by Kupiec Backtesting summarized that in this case, DGTN VaR provides the best risk assessment over different confidence levels (80, 90, 95, and 99 percent) compared to Delta Normal VaR and Delta Gamma Normal VaR.

Graph Convolutional Network with Syntactic Dependency for Aspect-Based Sentiment Analysis

AbstractAspect-based sentiment analysis (ABSA) aims to mine the sentiment tendencies expressed by specific aspect terms. The studies of ABSA mainly focus on the attention-based approaches and the graph neural network approaches based on dependency trees. However, the attention-based methods usually face difficulties in capturing long-distance syntactic dependencies. Additionally, existing approaches using graph neural networks have not made sufficient exploit the syntactic dependencies among aspects and opinions. In this paper, we propose a novel Syntactic Dependency Graph Convolutional Network (SD-GCN) model for ABSA. We employ the Biaffine Attention to model the sentence syntactic dependencies and build syntactic dependency graphs from aspects and emotional words. This allows our SD-GCN to learn both the semantic relationships of aspects and the overall semantic meaning. According to these graphs, the long-distance syntactic dependency relationships are captured by GCNs, which facilitates SD-GCN to capture the syntactic dependencies between aspects and viewpoints more comprehensively, and consequently yields enhanced aspect features. We conduct extensive experiments on four aspect-level sentiment datasets. The experimental results show that our SD-GCN outperforms other methodologies. Moreover, ablation experiments and visualization of attention further substantiate the effectiveness of SD-GCN.

Multi-tier scheduling algorithm of dispatching systems for urban water logging

Due to global warming, considerable amounts of storm rain have occurred, causing urban water logging and flooding. The efficient scheduling of drainage systems among pumping stations is crucial to mitigating flash flooding in urban areas. This study introduces a Multi-Level Dynamic Priority and Importance Scheduling (MDPIS) algorithm as a proactive solution for addressing urban flooding through the optimization of drainage system discharge capacities. The algorithm's robustness is guaranteed through the integration of a multi-tier drainage system and dependency relationships. Additionally, the incorporation of an importance parameter is considered for facilitating the practical exploration of flooding risk evaluation. The proposed model was applied to simulate a drainage system in Haining City, and the results indicate that its accuracy, flexibility and reliability outperform that of existing algorithms such as fixed-priority scheduling. Moreover, the proposed approach enabled a considerable reduction in overflow loss and improved the efficiency of the sewage system. This method can improve the responses of cities to the rising problem of urban water logging.

Understanding drought propagation through coupling spatiotemporal features using vine copulas: A compound drought perspective

Accurately evaluating drought impact on agriculture poses a challenge to regional food security, particularly in compound drought (i.e., meteorological and agricultural drought co-occurring) scenarios. This study presents a novel approach utilizing Vine copula for coupling spatiotemporal features to evaluate drought propagation. Three-dimensional clustering method was employed to identify meteorological and agricultural drought events, which excelled in capturing dynamic evolution characteristics (duration, area, severity, etc.) as well as integrating them into comprehensive meteorological drought intensity (IMD) and agricultural drought intensity (IAD). Through spatiotemporal matching, compound drought events were extracted from the meteorological-agricultural drought event pairs. From compound drought perspective, compound duration (CD) and compound area (CA) were devised to characterize drought propagation potential across time and space. Finally, the Vine copula method was employed to model the interdependence between four key coupling features, namely IMD, IAD, CD, and CA, and evaluate the probability of triggering agricultural drought with different intensity levels. Results showed that CD and CA can respectively characterize the temporal and spatial accumulation scale of drought propagation. At a certain IMD level, CD significantly influences the propagation probability (i.e., “stratification” phenomenon), while CA increases the probability proportionally. Probability evaluation lacking spatiotemporal information may underestimate the likelihood of drought propagation characterized by “low-IMD” but “long-CD” or “large-CA”. The four-dimensional Vine copula structure can effectively couple dependence relationships of compound drought characteristics, and exhibits reliable robustness. This research provides stakeholders accurate probabilistic evaluation under compound drought scenarios, offering new insight into drought propagation.

The civilizing process theory as a theoretical-methodological possibility in the humanities and social sciences

Norbert Elias studied the development processes of different occidental societies. Adopting long-term comparative empirical research, the author built a conceptual repertoire that gave bases to the Civilizing Process Theory. His findings demonstrated that there is a mutual dependence relationship between social development (sociogenesis) and changes in individual consciousness and self-control (psychogenesis), present in different phenomena of human socialization. In this sense, this essay aims to present the Civilizing Process Theory fundamental concepts, developed by Norbert Elias, trying to demonstrate its potential as a theoretical-methodological reference in the Human and Social Sciences researches.

A Literature Review of Seaport Decarbonisation: Solution Measures and Roadmap to Net Zero

This paper provides a comprehensive review of the literature related to seaport decarbonisation by combining the academic literature with case studies, industrial reports, newsletters, and domain knowledge. Through the literature review, the emission sources at seaports are categorised according to different criteria for better understanding. One of the criteria is the geographic location, which divides the emission sources into four categories. For each emission source category, the emission reduction measures in the literature are categorised into six structured categories including operational measures, technical measures, fuel and energy measures, infrastructural measures, digitalisation measures, and policy and collaboration measures. The first three categories have a direct impact on emission reductions, whereas the last three categories tend to support and facilitate the development and implementation of the first three categories. Representative case studies are selected from the UK ports to discuss their decarbonisation practices and pathways to net zero. We then propose a generic time-phased roadmap for port decarbonisation towards net zero, which divides the solution measures in each category into three phases to show their progressive processes. We explain the dependence relationships of the solution measures in the roadmap and discuss the challenges and opportunities in the implementation of the roadmap. This paper could offer strategic guidelines to port-associated stakeholders to implement emission reduction strategies and transition to net zero from the system perspective.

Exploring the relationship between intestinal microbiota and immune checkpoint inhibitors in the treatment of non-small cell lung cancer: insights from the "lung and large intestine stand in exterior-interior relationship" theory.

This study is aim to discern the Traditional Chinese Medicine (TCM) syndrome classifications relevant to immunotherapy sensitive in non-small cell lung cancer (NSCLC) patients, and to delineate intestinal microbiota biomarkers and impact that wield influence over the efficacy of NSCLC immunotherapy, grounded in the TCM theory of "lung and large intestine stand in exterior-interior relationship." The study cohort consisted of patients with advanced NSCLC who received treatment at the Oncology Department of Chengdu Fifth People's Hospital. These patients were categorized into distinct TCM syndrome types and subsequently administered immune checkpoint inhibitors (ICIs), specifically PD-1 inhibitors. Stool specimens were collected from patients both prior to and following treatment. To scrutinize the differences in microbial gene sequences and species of the intestinal microbiota, 16S rRNA amplicon sequencing technology was employed. Additionally, peripheral blood samples were collected, and the analysis encompassed the assessment of T lymphocyte subsets and myeloid suppressor cell subsets via flow cytometry. Subsequently, alterations in the immune microenvironment pre- and post-treatment were thoroughly analyzed. The predominant clinical manifestations of advanced NSCLC patients encompassed spleen-lung Qi deficiency syndrome and Qi-Yin deficiency syndrome. Notably, the latter exhibited enhanced responsiveness to ICIs with a discernible amelioration of the immune microenvironment. Following ICIs treatment, significant variations in microbial abundance were identified among the three strains: Clostridia, Lachnospiraceae, and Lachnospirales, with a mutual dependency relationship. In the subset of patients manifesting positive PD-L1 expression and enduring therapeutic benefits, the study recorded marked increases in the ratios of CD3+%, CD4+%, and CD4+/CD8+ within the T lymphocyte subsets. Conversely, reductions were observed in the ratios of CD8%, Treg/CD4+, M-MDSC/MDSC, and G-MDSC/MDSC. The strains Clostridia, Lachnospiraceae, and Lachnospirales emerge as potential biomarkers denoting the composition of the intestinal microbiota in the NSCLC therapy. The immunotherapy efficacy of ICIs markedly accentuates in patients displaying durable treatment benefits and those expressing positive PD-L1.

GRRLN: Gated Recurrent Residual Learning Networks for code clone detection

AbstractCode clone detection is a critical problem in software development and maintenance domains. It aims to identify functionally identical or similar code fragments within an application. Existing works formulate the code clone detection task as a binary classification problem which predicts a code pair as a clone or not based on a pre‐defined threshold. In reality, there are various types of code clone subject to the degree of how a pair of code fragments are similar to each other. To investigate the effect of different code clone detection manners on the clone detection result, we propose Gated Recurrent Residual Learning Networks (GRRLN), a novel neural network model for code clone detection. To train GRRLN, we first represent each code fragment as a statement‐level tree sequence derived from the whole abstract syntax tree (AST). Then, a gated recurrent neural network with residual connections is adopted to fully extract the semantics of all individual statement trees together with their dependency relationships across the input statement sequence. Finally, the output representations of code fragments by GRRLN are used for similarity calculation and clone detection. We evaluate GRRLN using two real‐world datasets for code clone detection and clone type classification. Experiments show that GRRLN achieves promising and compelling results and meanwhile needs significantly less time and memory consumption compared with the state‐of‐the‐art methods.

Causal modeling in multilevel settings: A new proposal

AbstractAn important question for the causal modeling approach is how to integrate non‐causal dependence relations such as asymmetric supervenience into the approach. The most prominent proposal to that effect (due to Gebharter) is to treat those dependence relationships as formally analogous to causal relationships. We argue that this proposal neglects some crucial differences between causal and non‐causal dependencies, and that in the context of causal modeling non‐causal dependence relationships should be represented as mutual dependence relationships. We develop a new kind of model – “hybrid models” ‐ based on this suggestion, and formulate a set of axioms for those models. Our formalism has important implications for Kim's exclusion problem: whereas Gebharter's framework vindicates Kim's causal exclusion objection against nonreductive physicalism, our framework has no such implication, and can help non‐reductive physicalists vindicate the efficacy of high‐level properties. A further benefit of our formalism is that it yields a natural and plausible way of thinking about interventions in multi‐level contexts.

Mixture of Attention Variants for Modal Fusion in Multi-Modal Sentiment Analysis

With the popularization of better network access and the penetration of personal smartphones in today’s world, the explosion of multi-modal data, particularly opinionated video messages, has created urgent demands and immense opportunities for Multi-Modal Sentiment Analysis (MSA). Deep learning with the attention mechanism has served as the foundation technique for most state-of-the-art MSA models due to its ability to learn complex inter- and intra-relationships among different modalities embedded in video messages, both temporally and spatially. However, modal fusion is still a major challenge due to the vast feature space created by the interactions among different data modalities. To address the modal fusion challenge, we propose an MSA algorithm based on deep learning and the attention mechanism, namely the Mixture of Attention Variants for Modal Fusion (MAVMF). The MAVMF algorithm includes a two-stage process: in stage one, self-attention is applied to effectively extract image and text features, and the dependency relationships in the context of video discourse are captured by a bidirectional gated recurrent neural module; in stage two, four multi-modal attention variants are leveraged to learn the emotional contributions of important features from different modalities. Our proposed approach is end-to-end and has been shown to achieve a superior performance to the state-of-the-art algorithms when tested with two largest public datasets, CMU-MOSI and CMU-MOSEI.