Use Of Local Information Research Articles

A frequency selection network for medical image segmentation

Existing medical image segmentation methods may only consider feature extraction and information processing in spatial domain, or lack the design of interaction between frequency information and spatial information, or ignore the semantic gaps between shallow and deep features, and lead to inaccurate segmentation results. Therefore, in this paper, we propose a novel frequency selection segmentation network (FSSN), which achieves more accurate lesion segmentation by fusing local spatial features and global frequency information, better design of feature interactions, and suppressing low correlation frequency components for mitigating semantic gaps. Firstly, we propose a global-local feature aggregation module (GLAM) to simultaneously capture multi-scale local features in the spatial domain and exploits global frequency information in the frequency domain, and achieves complementary fusion of local details features and global frequency information. Secondly, we propose a feature filter module (FFM) to mitigate semantic gaps when we conduct cross-level features fusion, and makes FSSN discriminatively determine which frequency information should be preserved for accurate lesion segmentation. Finally, in order to make better use of local information, especially the boundary of lesion region, we employ deformable convolution (DC) to extract pertinent features in the local range, and makes our FSSN can focus on relevant image contents better. Extensive experiments on two public benchmark datasets show that compared with representative medical image segmentation methods, our FSSN can obtain more accurate lesion segmentation results in terms of both objective evaluation indicators and subjective visual effects with fewer parameters and lower computational complexity.

The Avignon Chart (c. 1300–c. 1310): An Early Attempt to Represent the Northern Regions of Europe on a Nautical Chart

ABSTRACT The Avignon chart, which came to light in 2002, is one of the earliest surviving charts of the Mediterranean area. Despite its fragmentary nature and damage from reuse as a book cover in the sixteenth century, its precocity alone—it is thought to date from between c. 1300 and c. 1310, following only the Pisane (c. 1270) and the Cortona (c. 1300) charts—would be enough to ensure its importance. Researches to date, however, underline its particular significance in the history of medieval chart making. Pending further, multidisciplinary research, the present article draws attention to the main characteristics of the Avignon chart and highlights some of the enigmas of its construction, such as the way the chartmaker manipulated the orientation of the Iberian Peninsula in order to create space at the top to show, for the first time, the northern coastline of mainland Europe as far as the North and Baltic seas. The anonymous creator also constructed a grid to ensure the correct alignment of Britain, whose detailed configuration points to the use of local information among the available geographical sources for, and cultural knowledge about, the island.

MFF: An effective method of solving the ill regions in stereo matching

AbstractIn the current stereo matching field, the accuracy of the derived disparity map is highly dependent on the processing capability of ill regions. Fortunately, we find that the use of local information will eliminate the negative effects associated with ill regions. As a result of the above discovery, we propose the Concatenated Dilated Convolution (CDC) block and the Multi‐scale Feature Fusion module (MFF), which are capable of effectively extracting regional context by increasing the receptive field in parallel and channel‐wise ways. The CDC block can expand the receptive field by applying multiple dilated convolutions at different dilation rates in parallel to enhance the smoothness of the feature map. By constructing parallel CDC blocks in a multiple dilated manner, the MFF module can improve the smoothness of the feature map. In addition, to control the number of parameters in the MFF network, a high‐performance channel‐distribution algorithm is proposed, capable of adjusting the weights of each module and convolution in an adaptive manner while reducing the number of parameters. Extensive experiments have demonstrated that MFF and CDC can effectively improve the performance of ill areas and networks with a minimal number of parameters.

Designing robust scale-free networks under targeted link attack using local information

We study the effects of deliberate attacks to important links in scale-free networks and propose simple strategies to mitigate the damage. While strategies that require global network information and heavy computation have been proposed, here we focus on heuristics that make use of local information only and are much easier to compute. Using model scale-free networks and under the constraint of an invariant degree distribution, we show by numerical simulation that our approach in the average allows to enhance the network robustness notably, as measured by the integrity of the largest connected component. Because it is fast, the approach is also applicable to real-world networks of which we give two typical examples. Overall, the results are equivalent to those obtained in more complex settings that require global network knowledge and are much more computer-intensive, an important point for large networks.

High-dimensional nonlinear Bayesian inference of poroelastic fields from pressure data

We investigate solution methods for large-scale inverse problems governed by partial differential equations (PDEs) via Bayesian inference. The Bayesian framework provides a statistical setting to infer uncertain parameters from noisy measurements. To quantify posterior uncertainty, we adopt Markov Chain Monte Carlo (MCMC) approaches for generating samples. To increase the efficiency of these approaches in high-dimension, we make use of local information about gradient and Hessian of the target potential, also via Hamiltonian Monte Carlo (HMC). Our target application is inferring the field of soil permeability processing observations of pore pressure, using a nonlinear PDE poromechanics model for predicting pressure from permeability. We compare the performance of different sampling approaches in this and other settings. We also investigate the effect of dimensionality and non-gaussianity of distributions on the performance of different sampling methods.

Moisture as a Driver of Long-Term Threats to Timber Heritage—Part II: Risks Imposed on Structures at Local Sites

Timber heritage sites are vulnerable to damage from moisture. Simple meteorological descriptions of climate need to be tuned to capture drivers that threaten heritage, including dimensional change, insect attack and mould growth. Global climate models often provide projections through to the end of the 21st century but need to be translated to a local level to reveal processes of deterioration at specific sites. Translation to a local level can be challenging and requires the use of local information from a range of sources. This translation is explored over a range of sites facing different climate pressures, including fungal and insect risk at Harmondsworth Great Barn, England; changes in humidity range, salt risk and algal growth in rural timber buildings in the Midwestern states, USA; wind-driven rain impacts on board houses in Freetown, Sierra Leone; and rainfall and humidity range on timber buildings among the tropical rainforests of the Amazon, Congo Basin and Southeast Asia. Evidence-based narratives provide a tool to incorporate a multiplicity of local information to enrich projections and the interpretation of the model output. These could build trust and aid decision-making based on future projections, which are inherently uncertain.

Distributed Robust Dictionary Pair Learning and Its Application to Aluminum Electrolysis Industrial Process

In modern industrial systems, high-dimensional process data provide rich information for process monitoring. To make full use of local information of industrial process, a distributed robust dictionary pair learning (DRDPL) is proposed for refined process monitoring. Firstly, the global system is divided into several sub-blocks based on the reliable prior knowledge of industrial processes, which achieves dimensionality reduction and reduces process complexity. Secondly, a robust dictionary pair learning (RDPL) method is developed to build a local monitoring model for each sub-block. The sparse constraint with l2,1 norm is added to the analytical dictionary, and a low rank constraint is applied to the synthetical dictionary, so as to obtain robust dictionary pairs. Then, Bayesian inference method is introduced to fuse local monitoring information to global anomaly detection, and the block contribution index and variable contribution index are used to realize anomaly isolation. Finally, the effectiveness of the proposed method is verified by a numerical simulation experiment and Tennessee Eastman benchmark tests, and the proposed method is then successfully applied to a real-world aluminum electrolysis process.

Lightweight image super-resolution with multiscale residual attention network

In recent years, various convolutional neural networks have successfully applied to single-image super-resolution task. However, most existing models with deeper or wider networks require heavy computation and memory consumption that restrict them in practice. To solve the above questions, we propose a lightweight multiscale residual attention network, which not merely can extract more detail to improve the quality of the image but also decrease the usage of the parameters. More specifically, a multiscale residual attention block (MRAB) as the basic unit can fully exploit the image features with different sizes of convolutional kernels. Meanwhile, the attention mechanism can be adaptive to recalibrate channel and spatial information of feature mappings. Furthermore, a local information integration module (LFIM) is designed as the network architecture to maximize the use of local information. The LFIM consists of several MRAB and a local skip connection to complement information loss. Our experimental results show that our method is superior to the representative algorithms in performance with fewer parameters and computational overhead. Code is available at https://github.com/xiaotian3/EMRAB.

Emergence of adaptive global movement from a subjective inference about local resource distribution

In animal foraging, the optimal search strategy in an unknown environment varies depending on the context, such as the resource density and season. When food is distributed sparsely and uniformly, superdiffusive walks outperform normal-diffusive walks. However, superdiffusive walks are no longer advantageous when random walkers forage in resource-rich environments. It is not currently clear whether a relationship exists between an agent's use of local information to make subjective inferences about global food distribution and the optimal random walk strategy. Therefore, I investigated how flexible exploration is achieved if an agent alters its directional rule based on the local resource distribution. In the proposed model, the agent, a Brownian-like walker, estimates whether an abundant or sparse area is nearby using local resource patterns and then makes a decision by altering its movement rules. I show that the agent can behave like a non-Brownian walker if it interacts with a prey distribution. The agent can adaptively switch between diffusive properties depending on the resource density. This leads to a more effective resource-searching performance than a simple random-walk model. These results demonstrate that optimal searching is a context-dependent process.

Multi-Self-Attention for Aspect Category Detection and Biomedical Multilabel Text Classification with BERT

Multilabel classification is one of the most challenging tasks in natural language processing, posing greater technical difficulties than single-label classification. At the same time, multilabel classification has more natural applications. For individual labels, the whole piece of text has different focuses or component distributions, which require full use of local information of the sentence. As a widely adopted mechanism in natural language processing, attention becomes a natural choice for the issue. This paper proposes a multilayer self-attention model to deal with aspect category and word attention at different granularities. Combined with the BERT pretraining model, it achieves competitive performance in aspect category detection and electronic medical records’ classification.

Communication-less ensemble classifier-based protection scheme for DC microgrid with adaptiveness to network reconfiguration and weather intermittency

DC microgrids have received wide interest in recent times because of the advantages related to improved power transfer capability, high efficiency, compatibility with distributed energy resources (DER) and reduced losses. However, the adoption of DC microgrids has been restricted because of the challenges in designing a reliable protection scheme. The challenge results from the high magnitude of the fault current without zero crossing, coupled with the requirement of high rapidity in fault detection. The scenario is further complicated during the variation in operational dynamics of microgrid either due to weather intermittency or network reconfiguration arising out of DER outage. Non-addressal of the above issues results in reduced grid resilience because of possible relay malfunction during stressed scenarios. In this regard, an ensemble classifier-based protection scheme has been proposed for DC microgrid with adaptability to system reconfiguration and weather intermittency. The adaptiveness has been achieved by online identification of network topology, while immunity to weather variation is attained by stochastic modelling of solar irradiance and wind speed. The use of local information for executing the protection tasks avoids the issues related to data loss and latency in the communication link. The task of operating mode detection, topology identification, fault detection/classification and section identification have been formulated as a set of classification problems and further solved using a random subspace sampling-based ensemble classifier technique. The reliability of the proposed scheme has been extensively validated for a wide range of fault scenarios involving wide variation in network topology, weather condition and fault parameters.

Multiblock Dynamic Slow Feature Analysis-Based System Monitoring for Electrical Drives of High-Speed Trains

The electrical drive system of high-speed trains is a key subsystem to ensure the continuous supply of train power and stable operation. By the use of local information, this article presents a method called multiblock dynamic slow feature analysis (MBDSFA) with its application in the electrical drive system of high-speed trains. First, the relevance among all variables of electrical drive systems is calculated by using mutual information, based on which all variables are divided into blocks. Then, the dynamic slow feature analysis-based system monitoring scheme is employed for each subblock, and the local characteristics of electrical drive systems are analyzed via two kinds of test statistics. All subblocks are integrated based on the Bayesian inference to obtain the global monitoring results. Finally, the effectiveness and feasibility of the proposed approach are verified through the case study on the electrical drive system of high-speed trains.

A new attributed graph clustering by using label propagation in complex networks

The diffusion method is one of the main methods of community detection in complex networks. In this method, the use of the concept that diffusion within the nodes that are members of a community is faster than the diffusion of nodes that are not in the same community. In this way, the dense subgraph will detect the graph in the middle layer. The LPA algorithm, which mimics epidemic contagion by spreading labels, has attracted much attention in recent years as one of the most efficient algorithms in the subcategory of diffusion methods. This algorithm is one of the detection algorithms of most popular communities in recent years because of possessing some advantages including linear time order, the use of local information, and non-dependence on any parameter; however, due to the random behavior in LPA, there are some problems such as unstable and low quality resulting from larger monster communities. This algorithm is easily adaptable to attributed network. In this paper, it is supposed to propose a new version of the LPA algorithm for attributed graphs so that the detected communities solve the problems related to unstable and low quality in addition to possessing structural cohesiveness and attribute homogeneity. For this purpose, a weighted graph of the combination of node attributes and topological structure is produced from an attributed graph for nodes which have edges with each other. Also, the centrality of each node will be calculated equal to the influence of each node using Laplacian centrality, and the steps of selecting the node are being enhanced for updating as well as the mechanism of updating based on the influence of nodes. The proposed method has been compared to other primary and new attributed graph clustering algorithms for real and artificial datasets. In accordance with the results of the experiments on the proposed algorithm without parameter adjusting for different networks of density and entropy criteria, the normalized mutual information indicates that the proposed method is more efficient and precise than other state-of-the-art attributed graph clustering methods.

The use of local information sharing on soccer game optimization

Research that focuses on the performance improvement of metaheuristics is important to gain understanding on the characteristic of specific algorithm. Better understanding of metaheuristics method will bring benefit when it comes to implement the methods to solve real problems. This paper studies the effect of local information sharing on the performance of soccer game optimization with suitable control parameter settings. A novel method, called soccer game optimization with local information sharing (SGOLS), has been proposed in this paper. The method implements local information derived from several nearby players to conduct move forward. Therefore, the move forward will consider the position of the ball dribbler, the previous best player position as well as the position of players nearby. The proposed method is evaluated based on 20 unconstraint continuous problems, consisting of unimodal function and multimodal functions, and compared to SGO, PSO and DE. The experiment result reveals that the local information sharing could enhance intensification search in SGOLS. The proposed methods perform better in high-dimensionality problems than SGO, PSO and DE. However, the use of local information consumes more computational times than the SGO and PSO, but it is still faster than DE.

Consensus seeking for heterogeneous networks of agents with non‐convex constraints and switching topologies

This study focuses on the problem of constraint consensus of discrete-time multi-agent systems with high-order heterogeneity. It is assumed that the agents are of different orders in such systems ranging from order 1 to l and the velocity of each agent is constrained to lie in a non-convex set. This study proposes a distributed constraint control algorithm which guarantees the agents converge to a common point through the use of local information. Model transformations of the system matrix are performed in this study, so that the authors can utilise the properties of the stochastic matrix. Meanwhile, the boundness of state-dependent stochastic matrices and the convexity of stochastic matrices are analysed and explored in detail. Finally, a numerical example is given to illustrate the correctness of the results.

Probabilistic forecasts of time and energy flexibility in battery electric vehicle charging

Users charging the batteries of their electric vehicles in an uncoordinated manner can present energy systems with a challenge. One possible solution, smart charging, relies on the flexibility within each charging process and controls the charging process to optimize different objectives. Effective smart charging requires forecasts of energy requirements and parking duration at the charging station for each individual charging process. We use data from travel logs to create quantile forecasts of parking duration and energy requirements, approximated by upcoming trip distance. For this task, we apply quantile regression, multi-layer perceptrons with tilted loss function, and multivariate conditional kernel density estimators. The out-of-sample evaluation shows that the use of local information from the vehicle's travel data improves the forecasting accuracy by 13.7% for parking duration and 0.56% for trip distance compared to the data generated at the charging stations. In addition, the analysis of a case study shows that using probabilistic forecasts can control the interruption of charging processes more efficiently compared to point forecasts. Probabilistic forecasting leads up to 7.0% less interruptions, which can cause a restriction in drivers' mobility demand. The results show that charging station operators benefit from leveraging the driving patterns of electric vehicles. Thereby, smart charging and the application of the proposed models as benchmarks models for the related forecasting tasks is an improvement for the operators.

Deep Learning-Based Decoding of Constrained Sequence Codes

Constrained sequence (CS) codes, including fixed-length CS codes and variable-length CS codes, have been widely used in modern wireless communication and data storage systems. Sequences encoded with constrained sequence codes satisfy constraints imposed by the physical channel to enable efficient and reliable transmission of coded symbols. In this paper, we propose using deep learning approaches to decode fixed-length and variable-length CS codes. Traditional encoding and decoding of fixed-length CS codes rely on look-up tables (LUTs), which is prone to errors that occur during transmission. We introduce fixed-length constrained sequence decoding based on multiple layer perception (MLP) networks and convolutional neural networks (CNNs), and demonstrate that we are able to achieve low bit error rates that are close to maximum a posteriori probability (MAP) decoding as well as improve the system throughput. Further, implementation of capacity-achieving fixed-length codes, where the complexity is prohibitively high with LUT decoding, becomes practical with deep learning-based decoding. We then consider CNN-aided decoding of variable-length CS codes. Different from conventional decoding where the received sequence is processed bit-by-bit, we propose using CNNs to perform one-shot batch-processing of variable-length CS codes such that an entire batch is decoded at once, which improves the system throughput. Moreover, since the CNNs can exploit global information with batch-processing instead of only making use of local information as in conventional bit-by-bit processing, the error rates can be reduced. We present simulation results that show excellent performance with both fixed-length and variable-length CS codes that are used in the frontiers of wireless communication systems.

Distributed lifetime optimization in wireless sensor networks using alternating direction method of multipliers

SummaryDue to the limited energy of sensor nodes in wireless sensor networks, extending the network lifetime is a major challenge that can be formulated as an optimization problem. In this paper, we propose a distributed iterative algorithm based on alternating direction method of multipliers with the aim of maximizing sensor network lifetime. The features of this algorithm are the use of local information, low overhead of message passing, low computational complexity, fast convergence, and, consequently, reduced energy consumption. In this study, we present the convergence results and the number of iterations required to achieve the stopping criterion. Furthermore, the impact of problem size (number of sensor nodes) on the solution and constraints violation is studied, and, finally, the proposed algorithm is compared with one of the well‐known subgradient‐based algorithms.

Remote sensing image segmentation based on a robust fuzzy C-means algorithm improved by a parallel Lévy grey wolf algorithm.

Due to the insufficient use of local information, the traditional fuzzy C-means (FCM) algorithm and its extension algorithm combined with spatial information show poor robustness and low segmentation accuracy. In addition, in the process of image segmentation based on the FCM algorithm, the initial center estimation is regarded as the process of searching the appropriate value in the gray range. To solve these problems, a new robust algorithm is proposed in this paper. The algorithm searches the optimal initial center by introducing an improved parallel Lévy grey wolf optimization algorithm, which is an improved fuzzy C-means segmentation algorithm that combines local information and adaptive gray weighting. Experimental results infer that both the precision and efficiency of the proposed method are superior to those of the state-of-arts.

Integration of Local and Global Support Vector Machines to Improve Urban Growth Modelling

The use of local information for the classification and modelling of spatial variables has increased with the application of statistical and machine learning algorithms, such as support vector machines (SVMs). This study presents a new local SVM (LSVM) model that was developed to model the probability of urban development and simulate urban growth in a subregion in the southwestern suburb of the Tehran metropolitan area, Iran, for the periods of 1992–1996 and 1996–2002. Based on the focal training sample, the model was calibrated using the cross-validation method, and the optimal bandwidth was determined. The results were compared with those of a nonlinear global SVM (GSVM) model that was calibrated based on the ten-fold cross-validation method. This study then evaluated an integrated SVM model (LGSVM) obtained based on a weighted combination of the local and global urban development probabilities. A comparison of the probability maps showed a higher accuracy for the LGSVM than for either the LSVM or GSVM model. To assess the performance of the LSVM, GSVM and LGSVM models in the simulation of urban growth, probability maps were employed as the transition rules for urban cellular automata. The results show that a trade-off between local and global SVM models can enhance the performance of urban growth modelling.