Addressing the computational task offloading chal-lenges, this study develops a cloud-edge cooperative network framework comprising three layers: the user layer, edge layer, and cloud layer. To mitigate the issue of delayed rewards during training, a fixed-time step approach is adopted. In response to the limited robustness of single-agent reinforcement learning within cloud-edge collaborative settings, we introduce a multi-agent task offloading algorithm, designated as KLMAC-PPO, and detail its network architecture. Furthermore, we incorporate LSTM networks to enhance the prediction and management of device mobility. The proposed algorithm utilizes a KAN to derive an optimal offloading strategy. Experimental evaluations demonstrate that the distributed offloading approach significantly ameliorates task delay, reduces energy consumption, and lowers the drop rate under high-load scenarios.

Due to limited features and dense object layouts, object detection in UAV images is challenging. Given that existing feature fusion methods have not fully explored the relationship between fine- and coarse-grained features, direct feature fusion can result in poor correlation between them, hindering the representative capability of fine-grained semantic information. To alleviate this issue, we introduce a method of Coarse-fine Feature Interaction Alignment (CFIA), which enhances the correlation between coarse-grained and fine-grained features across multi-scale feature maps through their interactive alignment. Firstly, we present the Wavelet-based High-frequency Preserving Down-sampling (WHPD), utilizing wavelet transform to extract high-frequency information to enhance object boundaries, minimizing crucial fine-grained information loss. Secondly, we propose the Feature Refinement and Interaction Alignment Strategy (FRIAS), which achieves feature interaction alignment by establishing the association of feature maps between coarse-grained and fine-grained features. This enhances the representative capability of feature maps at various scales for detecting small objects. Extensive experiments on the VisDrone, CARPK, and Drone-vs-Bird datasets have demonstrated the effectiveness of the CFIA method, which is highly competitive with state-of-the-art methods. The code is available at https://github.com/b-yanchao/CFIA.git.

Weakly-supervised semantic segmentation (WSSS) aims to achieve pixel-level classification under image-level supervision. Recent class activation map (CAM)-based methods seek to expand foreground activation while suppressing background. However, they often overlook the uncertainty of CAM, where non-salient activation in some regions complicates semantic classification. These regions are typically dismissed as noise, resulting in inappropriate activations due to inadequate regularization. To resolve this, we introduce a Spatial Frequency-Aware Self-Distillation strategy (SFS). Firstly, to enhance the perception of high-frequency spatial information in uncertain regions, we propose a boundary self-distillation and uncertain region reconstruction strategy, which captures high-frequency boundary information and fine-grained spatial context in these regions. Secondly, to enhance the discrimination of low-frequency semantic features, we propose a contrastive attention mechanism that guides the Vision Transformer (ViT) to focus more on the foreground, thereby improving the distinction between foreground and background. Finally, our SFS demonstrates outstanding performance on both the VOC 2012 and COCO 2014 datasets, attributed to its superior spatial frequency perception capabilities. The code is available at https://github.com/fjoybest/SFS.

Current feature fusion strategies often fail to adequately account for the influence of activation intensity across different scales on small object features, which impedes the effective detection of small objects. To address this limitation, we propose the Region-Adaptive Feature Disentanglement and Enhancement (RAFDE) strategy, which improves both downsampling and feature fusion by leveraging activation intensity variations at multiple scales. First, we introduce the Boundary Transitional Region-enhanced Downsampling (BTRD) module, which enhances boundary transitional regions containing both strongly and weakly activated features, thereby mitigating the loss of crucial boundary information for small objects. Second, we present the Regional-Adaptive Feature Fusion (RAFF) module, which adaptively disentangles and fuses co-activated and uni-activated regions from adjacent levels into the current level, effectively reducing the risk of small objects being overwhelmed. Extensive experiments on several public datasets demonstrate that the RAFDE strategy is highly effective and outperforms state-of-the-art methods. The code is available at https://github.com/b-yanchao/RAFDE.git.

Recent advances in Weakly Supervised Semantic Segmentation (WSSS) focus on generating high-quality Class Activation Maps (CAMs) using image-level labels. However, the co-occurrence of foreground-background concepts in a single image often induces semantic confusion, which degrades the quality of conventional CAM-based approaches. In this paper, we propose VLHP, a novel framework that leverages vision-language hybrid prototypes to overcome semantic confusion. Specifically, VLHP constructs hybrid prototypes through cross-modal association between textual embeddings and visual features, generating discriminative semantic representations while effectively bridging the modality gap. To further improve discriminability, we introduce two dedicated strategies: Discriminative Explicit Alignment (DEA) to explore cross-modal consistent discrimination and Confounding Background Decoupling (CBD) to model co-occurring backgrounds and decouple them. Finally, a Prototype-driven Class-aware Decoder (PCD) employs these refined prototypes as category-specific priors to generate precise segmentation masks in a single-stage framework. Extensive experiments on PASCAL VOC and MS COCO benchmarks demonstrate that VLHP outperforms state-of-the-art alternatives. The code is available at https://github.com/fjy0105/VLHP.

With the application of multisource information in different fields, the combination of different types of information, such as numerical data and text information, has become a favourable choice for performing stock market analyses. Despite the rich information provided by multisource data, building structured relationships remains challenging. In addition, some market relationship-based analysis methods use a predefined graph structure as a stock relationship graph, which makes it impossible to sensitively aggregate attribute features, and these methods cannot dynamically update market relationships or relationship strengths. In this paper, we propose a novel dynamic attribute-driven graph attention network incorporating sentiment (AGATS) information, transaction data, and text data. Inspired by behavioural finance , we separately extract sentiment information as a factor of technical indicators, and further realize the early fusion of technical indicators and textual data through tensor fusion. In particular, real-time intramarket dependencies and key attribute information are captured with graph networks, enabling dynamic relationship and relationship strength updates. Experiments conducted on real datasets show that our model is capable of ourperforming previously developed methods in prediction and trading.

Graphical models have been used extensively for modeling brain connectivity networks. However, unmeasured confounders and correlations among measurements are often overlooked during model fitting, which may lead to spurious scientific discoveries. Motivated by functional magnetic resonance imaging (fMRI) studies, we propose a novel method for constructing brain connectivity networks with correlated replicates and latent effects. In a typical fMRI study, each participant is scanned and fMRI measurements are collected across a period of time. In many cases, subjects may have different states of mind that cannot be measured during the brain scan: for instance, some subjects may be awake during the first half of the brain scan, and may fall asleep during the second half of the brain scan. To model the correlation among replicates and latent effects induced by the different states of mind, we assume that the correlated replicates within each independent subject follow a one-lag vector autoregressive model, and that the latent effects induced by the unmeasured confounders are piecewise constant. Theoretical guarantees are established for parameter estimation. We demonstrate via extensive numerical studies that our method is able to estimate latent variable graphical models with correlated replicates more accurately than existing methods.

High-dimensional healthcare data, such as electronic health records (EHR) data and claims data, present two primary challenges due to the large number of variables and the need to consolidate data from multiple clinical sites. The third key challenge is the potential existence of heterogeneity in terms of covariate shift. In this paper, we propose a distributed learning algorithm accounting for covariate shift to estimate the average treatment effect (ATE) for high-dimensional data, named DisC2o-HD. Leveraging the surrogate likelihood method, our method calibrates the estimates of the propensity score and outcome models to approximately attain the desired covariate balancing property, while accounting for the covariate shift across multiple clinical sites. We show that our distributed covariate balancing propensity score estimator can approximate the pooled estimator, which is obtained by pooling the data from multiple sites together. The proposed estimator remains consistent if either the propensity score model or the outcome regression model is correctly specified. The semiparametric efficiency bound is achieved when both the propensity score and the outcome models are correctly specified. We conduct simulation studies to demonstrate the performance of the proposed algorithm; additionally, we conduct an empirical study to present the readiness of implementation and validity.

Deep learning models applied to medical image segmentation have achieved remarkable performance across various tasks. Despite their high accuracy, these models may produce predictions that clinicians deem anatomically implausible. This limitation arises due to the inherent anatomical variability and indistinct boundaries present in clinical segmentation targets, impeding the effectiveness of existing research efforts. To overcome these challenges, we propose an innovative architectural framework known as Densely Cross-scale Attention (DCA) network. This framework efficiently incorporates multi-scale local and global attention dependencies using dense connections, resulting in anatomically-plausible segmentations. DCA comprises three new modules: 1) a multi-level feature aggregation module that merges multi-scale local features to acquire feature representations with diverse granularities, thereby compensating for information loss caused by gradient dispersion; 2) a cross-scale attention module that facilitates the modeling of long-range dependencies across scales while effectively diminishing task-irrelevant noise; and 3) a multi-scale self-attention module that captures the multi-scale global spatial relationships of single-scale local features, further enhancing the model’s resilience in addressing complex lesions. The proposed efficient dense connections offer a versatile and potent methodology for modeling long-range dependencies across scales in any segmentation network. The efficacy of our approach is demonstrated through experiments conducted on two clinically relevant datasets, substantiating its ability to achieve superior segmentation accuracy and anatomical plausibility.

Transformer architecture has emerged to be successful in many natural language processing tasks. However, its applications to clinical practice remain largely unexplored. In this study, we propose a Robust Cross-Scale Hybrid Transformer (RCSHT) architecture for medical image segmentation, which can effectively enhance the multi-scale feature representations while integrating local features with global dependencies. Specifically, we propose two new modules based on the self-attention mechanism: PSCM and PCCM, which perform matrix low-rank transformation on cross-scale images in spatial and channel space, respectively, to effectively enhance the discriminant ability of long-range dependencies of multi-scale images and effectively reduce the computational complexity. Meanwhile, we use spatial positional encoding to perform spatial regularization on the two attention modules, so that the model has more opportunities to obtain valuable spatial details and further enhance the spatial retention ability. Besides, by fusing outputs of the two modules, the robustness and discriminant ability of the feature representations are further improved. Note that our hybrid architecture allows transformers to be initialized as convolutional networks without pre-training. Extensive experiments conducted on our newly proposed dataset have demonstrated our RCSHT notably outperforms the state-of-the-art methods by a large margin, holding the promise to generalize well on other medical image segmentation.

Single image deraining is an important problem in many computer vision tasks because rain streaks can severely degrade the image quality. Recently, deep convolution neural network (CNN) based single image deraining methods have been developed with encouraging performance. However, most of these algorithms are designed by stacking convolutional layers, which encounter obstacles in learning abstract feature representation effectively and can only obtain limited features in the local region. In this paper, we propose a recurrent multi-connection fusion network (RMCFN) to remove rain streaks from single images. Specifically, the RMCFN employs two key components and multiple connections to fully utilize and transfer features. Firstly, we use a multi-scale fusion memory block (MFMB) to exploit multi-scale features and obtain long-range dependencies, which is beneficial to feed useful information to a later stage. Moreover, to efficiently capture the informative features on the transmission, we fuse the features of different levels and employ a multi-connection manner to use the information within and between stages. Finally, we develop a dual attention enhancement block (DAEB) to explore the valuable channel and spatial components and only pass further useful features. Extensive experiments verify the superiority of our method in visual effect and quantitative results compared to the state-of-the-arts.

With the increasing adoption of electronic health records, there is an increasing interest in developing individualized treatment rules, which recommend treatments according to patients' characteristics, from large observational data. However, there is a lack of valid inference procedures for such rules developed from this type of data in the presence of high-dimensional covariates. In this work, we develop a penalized doubly robust method to estimate the optimal individualized treatment rule from high-dimensional data. We propose a split-and-pooled de-correlated score to construct hypothesis tests and confidence intervals. Our proposal adopts the data splitting to conquer the slow convergence rate of nuisance parameter estimations, such as non-parametric methods for outcome regression or propensity models. We establish the limiting distributions of the split-and-pooled de-correlated score test and the corresponding one-step estimator in high-dimensional setting. Simulation and real data analysis are conducted to demonstrate the superiority of the proposed method.

Regression Discontinuity (RD) design is commonly used to estimate the causal effect of a policy. Existing RD relies on the continuity assumption of potential outcomes. However, self selection leads to different distributions of covariates on two sides of the policy intervention, which violates this assumption. The standard RD estimators are no longer applicable in such setting. We show that the direct causal effect can still be recovered under a class of weighted average treatment effects. We propose a set of estimators through a weighted local linear regression framework and prove the consistency and asymptotic normality of the estimators. We apply our method to a novel data set from Microsoft Bing on Generalized Second Price (GSP) auction and show that by placing the advertisement on the second ranked position can increase the click-ability by 1.91%.

In clinical practice, as a powerful and independent risk indicator of cardiovascular disease (CVD), accurate coronary artery calcium (CAC) segmentation can provide important information for the early diagnosis of CVD. However, due to the small and inconsistent CAC usually has fuzzy boundaries, which leads existing segmentation methods to suffer from unsatisfactory performance. To tackle this challenge, we propose a novel Efficient Multi-scale Vision Transformers for CAC segmentation (CAC-EMVT), which uses both the local and global branches to jointly model short- and long-range dependencies. CAC-EMVT is mainly composed of three modules: 1) a key factor sampling (KFS) module, which is used to mine the key factors of the image to perform low-rank reconstruction of highly structured features; 2) a non-local sparse context fusion (NSCF) module, which is used to efficiently model the global context information of shallow texture features; and 3) a non-local multi-scale context aggregation (NMCA) module, which can be applied to cross-level features to collect long-range dependencies from multiple scales. Undeniably, the newly proposed decomposable positional encoding plays a vital role in the performance improvement of the above modules. Extensive experiments are conducted on the CT scans of 130 CVD patients under 4-fold cross-validation and have demonstrated our CAC-EMVT notably outperforms the state-of-the-art methods in terms of both the mean Dice similarity coefficient (mDice) of 75.39%± 3.17 and mean surface distance (MSD) of 1.93%± 0.46. This reveals the effectiveness and the potential of our model in the clinical setting.

Abstract Coronary artery calcium is a strong and independent marker of atherosclerosis and cardiovascular disease. Typically, the accurate segmentation of computed tomography images of the chest is an important prerequisite and basis for coronary artery calcium identification and analysis. However, this is very challenging in practice because the boundaries of coronary artery calcium, the small lesions with large shape variation, are very blurry, resulting in poor performance in existing studies. To tackle this challenge, we present a novel Attention‐based Multi‐Scale Network called AMSN, which can process information through both the main and boundary branches in parallel. Key to our AMSN is a new non‐local multi‐scale context encoder module, which is mainly composed of the multi‐scale attention mechanism and local global long short‐term memory module. By aggregating the multi‐scale context information, i.e. high‐resolution low‐level and low‐resolution high‐level features, the model's feature representative capability and deployment ability are improved effectively. Besides, we introduce a new boundary preserving loss, which can consider the boundary information of all coronary artery calcium together and establish links for the segmentation of different coronary artery calcium simultaneously. Extensive experiments demonstrate our AMSN enables reliable accurate coronary artery calcium segmentation for assisted cardiovascular disease diagnosis clinically.

Recently smoothing deep neural network based classifiers via isotropic Gaussian perturbation is shown to be an effective and scalable way to provide state-of-the-art probabilistic robustness guarantee against ℓ2 norm bounded adversarial perturbations. However, how to train a good base classifier that is accurate and robust when smoothed has not been fully investigated. In this work, we derive a new regularized risk, in which the regularizer can adaptively encourage the accuracy and robustness of the smoothed counterpart when training the base classifier. It is computationally efficient and can be implemented in parallel with other empirical defense methods. We discuss how to implement it under both standard (non-adversarial) and adversarial training scheme. At the same time, we also design a new certification algorithm, which can leverage the regularization effect to provide tighter robustness lower bound that holds with high probability. Our extensive experimentation demonstrates the effectiveness of the proposed training and certification approaches on CIFAR-10 and ImageNet datasets.

Automated pancreas segmentation in abdominal computed tomography (CT) scans is of high clinical relevance (i.e. pancreas cancer diagnosis and prognosis), but extremely difficult because the pancreas is a soft, small, and flexible abdominal organ with high anatomical variability, which causes the previous segmentation methods to result in low precision. In this study, the authors present a new deep recurrent adversarial network (DRAN) to tackle this challenge. DRAN contains three steps: (i) preserving global resolution of CT scans and modifying the receptive field of kernel adaptively through a dilated convolution autoencoder module; (ii) modelling contextual spatial correlation between neighbouring CT scan patches benefits from a specially designed local long short‐term memory module; and (iii) improving the performance and generalisation by leveraging an adversarial module, which can constrain the spatial smoothness consistency between continuous CT scans based on the long‐range spatial interaction. The system is evaluated on a dataset of 80 manually segmented CT volumes, using four‐fold cross‐validation. Its performance surpasses other state‐of‐the‐art methods, with the Dice similarity coefficient of and pixel‐wise accuracy of . Also, they perform a qualitative evaluation by an expert further revealing the effectiveness and potential of their DRAN as a clinical segmentation tool.

Motivated by modern applications in which one constructs graphical models based on a very large number of features, this paper introduces a new class of cluster-based graphical models, in which variable clustering is applied as an initial step for reducing the dimension of the feature space. We employ model assisted clustering, in which the clusters contain features that are similar to the same unobserved latent variable. Two different cluster-based Gaussian graphical models are considered: the latent variable graph, corresponding to the graphical model associated with the unobserved latent variables, and the cluster-average graph, corresponding to the vector of features averaged over clusters. Our study reveals that likelihood based inference for the latent graph, not analyzed previously, is analytically intractable. Our main contribution is the development and analysis of alternative estimation and inference strategies, for the precision matrix of an unobservable latent vector Z. We replace the likelihood of the data by an appropriate class of empirical risk functions, that can be specialized to the latent graphical model and to the simpler, but under-analyzed, cluster-average graphical model. The estimators thus derived can be used for inference on the graph structure, for instance on edge strength or pattern recovery. Inference is based on the asymptotic limits of the entry-wise estimates of the precision matrices associated with the conditional independence graphs under consideration. While taking the uncertainty induced by the clustering step into account, we establish Berry-Esseen central limit theorems for the proposed estimators. It is noteworthy that, although the clusters are estimated adaptively from the data, the central limit theorems regarding the entries of the estimated graphs are proved under the same conditions one would use if the clusters were known in advance. As an illustration of the usage of these newly developed inferential tools, we show that they can be reliably used for recovery of the sparsity pattern of the graphs we study, under FDR control, which is verified via simulation studies and an fMRI data analysis. These experimental results confirm the theoretically established difference between the two graph structures. Furthermore, the data analysis suggests that the latent variable graph, corresponding to the unobserved cluster centers, can help provide more insight into the understanding of the brain connectivity networks relative to the simpler, average-based, graph.

We consider parameter estimation and statistical inference of high-dimensional undirected graphical models for mixed data comprising both ordinal and continuous variables. We propose a flexible model called Latent Mixed Gaussian Copula Model that simultaneously deals with such mixed data by assuming that the observed ordinal variables are generated by latent variables. For parameter estimation, we introduce a convenient rank-based ensemble approach to estimate the latent correlation matrix, which can be subsequently applied to recover the latent graph structure. In addition, based on the ensemble estimator, we develop test statistics via a pseudo-likelihood approach to quantify the uncertainty associated with the low dimensional components of high-dimensional parameters. Our theoretical analysis shows the consistency of the estimator and asymptotic normality of the test statistic. Experiments on simulated and real gene expression data are conducted to validate our approach.

Individualized treatment rules aim to identify if, when, which, and to whom treatment should be applied. A globally aging population, rising healthcare costs, and increased access to patient-level data have created an urgent need for high-quality estimators of individualized treatment rules that can be applied to observational data. A recent and promising line of research for estimating individualized treatment rules recasts the problem of estimating an optimal treatment rule as a weighted classification problem. We consider a class of estimators for optimal treatment rules that are analogous to convex large-margin classifiers. The proposed class applies to observational data and is doubly-robust in the sense that correct specification of either a propensity or outcome model leads to consistent estimation of the optimal individualized treatment rule. Using techniques from semiparametric efficiency theory, we derive rates of convergence for the proposed estimators and use these rates to characterize the bias-variance trade-off for estimating individualized treatment rules with classification-based methods. Simulation experiments informed by these results demonstrate that it is possible to construct new estimators within the proposed framework that significantly outperform existing ones. We illustrate the proposed methods using data from a labor training program and a study of inflammatory bowel syndrome.

Forecasting the inflow and outflow of crowds at metro station, immediately controlling the number of people entering at some special times and places to avoid the occurrence of malignant events for public safety, is of great significance to subway stations management and very challenging as it is affected by many complex elements, such as inter region station flows, major events or activities, and weather. We propose a approach based on deep residual learning, called ST-DRN, to discern the pattern of spatial and temporal and integrally predict the inflow and outflow of crowds in each subway station of a city. We propose an end-to-end structure of ST-DRN based on distinct attributes of spatio-temporal data. More specifically, we apply the residual neural network framework to model the temporal nearby, day, and week properties of crowd in subway station. For each feature, we design a branch of residual convolutional units, each of which handles the spatial properties of subway crowd. ST-DRN learns to dynamically summation the output of the three residual neural networks, assigning different weights to each branch. The summation is also further combined with external elements, such as weather, holiday and workdays or weekends, to forecast the final traffic flow of crowds in each station. Evaluations on the automatic fare collection (AFC) system subway record data in Suzhou demonstrate that we proposed ST-DRN outperforms than three prominent baseline methods.

Summary Characterization of rock properties is vital in producing oil and gas from shale reservoirs in an economically viable fashion. The nano‐pore structure and ultralow permeability in shale reservoirs present challenges to the traditional experimental characterization methods. Digital rock physics for the estimation of rock properties, especially for shale reservoirs, has become a powerful tool that greatly complements to lab experiments by combining advance imaging techniques with numerical simulations. The lattice Boltzmann method (LBM) is a well‐applied numerical method to simulate the fluid flow in pore structures at multiple length scales. Usually, the LBM simulation is resource intense because of its computation complexity and is facing great numerical challenges in extremely large‐cale computation. In this paper, we propose a multi‐GPU parallel implementation of 3D LBM on a hybrid high‐performance computing cluster to perform large‐scale simulations in reconstructed digital rocks. The program provides multiscale solution, pore scale and representative elementary volume (REV) scale based on the resolution of digital rock images. Optimization strategies are applied on partitioning simulation domain, improving data communication efficiency and maximizing CUDA occupancy. When running on a cluster of 32 GPUs, the proposed parallel implementation achieves a speedup of 1074x comparing to the in‐house sequential program.

We propose a new class of semiparametric exponential family graphical models for the analysis of high dimensional mixed data. Different from the existing mixed graphical models, we allow the nodewise conditional distributions to be semiparametric generalized linear models with unspecified base measure functions. Thus, one advantage of our method is that it is unnecessary to specify the type of each node and the method is more convenient to apply in practice. Under the proposed model, we consider both problems of parameter estimation and hypothesis testing in high dimensions. In particular, we propose a symmetric pairwise score test for the presence of a single edge in the graph. Compared to the existing methods for hypothesis tests, our approach takes into account of the symmetry of the parameters, such that the inferential results are invariant with respect to the different parametrizations of the same edge. Thorough numerical simulations and a real data example are provided to back up our theoretical results.

We provide a general theory of the expectation-maximization (EM) algorithm for inferring high dimensional latent variable models. In particular, we make two contributions: (i) For parameter estimation, we propose a novel high dimensional EM algorithm which naturally incorporates sparsity structure into parameter estimation. With an appropriate initialization, this algorithm converges at a geometric rate and attains an estimator with the (near-)optimal statistical rate of convergence. (ii) Based on the obtained estimator, we propose a new inferential procedure for testing hypotheses for low dimensional components of high dimensional parameters. For a broad family of statistical models, our framework establishes the first computationally feasible approach for optimal estimation and asymptotic inference in high dimensions.

In this paper, we propose a new cooperative random access strategy for future cooperative networks. This scheme considers the different requirements of services, and exploits the cooperative relaying nature for multi-channel environments, e.g. orthogonal frequency division multiplexing access (OFDMA). Due to differentiation of services and definition of special channels, those users with real-time (RT) request can reserve the access channels in advance, while others with non-real-time (NRT) services can access to the BS or the nearest distributed relay stations (DRSs), which are introduced for cooperative relaying, through sharing those channels remained. The analyses and numerical results demonstrate that our scheme can achieve high throughput, low collision probability and low access delay compared with conventional slotted Aloha.