Neighborhood Structure Research Articles

Probabilistic Context Neighborhood model for lattices

We present the Probabilistic Context Neighborhood model designed for two-dimensional lattices as a variation of a Markov random field assuming discrete values. In this model, the neighborhood structure has a fixed geometry but a variable order, depending on the neighbors’ values. Our model extends the Probabilistic Context Tree model, originally applicable to one-dimensional space. It retains advantageous properties, such as representing the dependence neighborhood structure as a graph in a tree format, facilitating an understanding of model complexity. Furthermore, we adapt the algorithm used to estimate the Probabilistic Context Tree to estimate the parameters of the proposed model. We illustrate the accuracy of our estimation methodology through simulation studies. Additionally, we apply the Probabilistic Context Neighborhood model to spatial real-world data, showcasing its practical utility.

Not so Simple: Examining the Gendered Nature of Intimate Partner Assault Victimizations.

Although prior research has established a relationship between neighborhood structure and intimate partner violence (IPV), much of this literature focuses on serious IPV, despite the fact that less serious forms of IPV comprise the majority of IPV instances. Further, women are far more likely to experience all types of IPV victimization relative to men, especially simple assault, yet the relative importance of neighborhood structure across gradations of IPV and victim sex is relatively unexplored. We use data from Los Angeles, CA, and disaggregate IPV assaults across victim sex to examine whether neighborhood factors have differential effects on simple and aggravated IPV. We find differential effects of neighborhood structure by victim sex, especially for simple assault. Most notably, we find that neighborhood racial composition has significantly greater effects on females relative to male victims of simple IPV assault, while residential stability is protective of women more so than men. Only the percent of vacant or owner-occupied housing was specific to male simple IPV victimization. This suggests that neighborhood-based interventions should differ depending on the type of IPV being targeted.

Geometric batch optimization for packing equal circles in a circle on large scale

The problem of packing equal circles in a circular container is a classic and famous packing problem, which is well-studied in academia and has a variety of applications in industry. The problem is computationally challenging, and researchers mainly focus on small to moderate-scale instances with the number of circular items n less than 320 in the literature. In this work, we aim to solve this problem on large scale, and it is the bottleneck for most global optimization problems. Specifically, we propose a novel geometric batch optimization method that makes batch gradient descent based on the geometric locations of packing items. This method not only can speed up the convergence process of continuous optimization significantly but also reduce the memory requirement during the program’s runtime. Then, we propose a heuristic search method, called solution-space exploring and descent, that can discover a feasible solution efficiently on large scale. Besides, we propose an adaptive neighbor object maintenance method to maintain the neighbor structure applied in the continuous optimization process. In this way, we can find high-quality solutions on large-scale instances within reasonable computational times. Extensive experiments on the benchmark instances sampled from n= 300 to 1000 show that our proposed algorithm outperforms the state-of-the-art algorithms and performs excellently on large-scale instances. In particular, our algorithm found 10 improved solutions among the 21 well-studied moderate-scale instances and 95 improved solutions among the 101 sampled large-scale instances. Furthermore, our geometric batch optimization, heuristic search, and adaptive maintenance methods are general and can be adapted to other packing and continuous optimization problems.

Cloud-Edge Test-Time Adaptation for Cross-Domain Online Machinery Fault Diagnosis via Customized Contrastive Learning

Nowadays offline transfer learning (TL) is the mainstream research for cross-domain machinery fault diagnosis (MFD). However, the target data is usually collected online by edge devices in real-world applications, causing practicality issue of offline TL. To address this issue, a new and practical online TL scenario test-time adaptation (TTA) is considered in this paper. In TTA, the labeled source data is inaccessible for privacy, and only a pre-trained source model is provided for online model adaptation with the online unlabeled target data. To enable TTA for MFD, a dynamic data division strategy is proposed to divide each mini-batch of the online target data into the certain-aware and uncertain-aware sets based on the spectral entropy and prediction confidence for the subsequent fine-grained online model adaptation. Delving into underlying data properties, a customized contrastive learning (CL) framework is proposed, consisting of specifically designed CL strategies for the different divided sets, respectively. Specifically, prototypical CL based on the representative class-wise prototypes is proposed to boost the feature discriminability of certain-aware set, while neighborhood CL based on the local neighborhood structure is proposed to refine the numerous noise of uncertain-aware set. Meanwhile, smooth entropy minimization is devised for reliable model uncertainty reduction. The cloud-edge TTA implementation framework is further considered in the scenario of cloud manufacturing and edge computing for practical real-world applications. Specifically, the pre-trained source model is directly deployed from cloud to edge for local TTA. A simple source model pruning strategy is proposed to obtain the lightweight pre-trained source model for efficient deployment on resource-limited edge devices. Practical experiments on two edge devices Raspberry Pi and NVIDIA Jetson Nano verify the effectiveness and efficiency.

The effect of surgeon’s pose and duration of operation on movement accuracy and level of microsurgical hand tremor

Background. Physiological hand tremor is one of significant problems in microsurgical technique. One careless movement can cause damage to the neighboring structures and consequently worsen a patient’s condition. Therefore, the problem of reducing hand tremor remains relevant.Aim. To determine the effect of microsurgeon’s pose, duration of operation on the accuracy of surgical action and level of microsurgical hand tremor.Materials and methods. The study included 14 neurosurgery residents satisfying inclusion criteria. For evaluation and simulation of microsurgical action, validated and developed by the study authors devices for testing and surgical microscope Carl Zeiss were used. The experimental results were compared using univariate analysis. The differences were considered statistically significant at p <0.05.Results. Surgeon’s pose standing / sitting did not affect spatial accuracy of surgical action and the level of microsurgical hand tremor. However, statistically significant changes in hand tremor were observed both in standing and sitting positions with increased time of microsurgical action.Conclusion. The choice between sitting or standing pose for microsurgical operation is not a factor significantly affecting microsurgeon’s spatial accuracy of movement and hand tremor. Duration of surgical intervention affects the level of microsurgical hand tremor.

基于改进鲸鱼优化算法的多农业搬运机器人协同作业轨迹规划研究

To solve the problem that the moving trajectory and operating trajectory are relatively independent and time-consuming when robots transfer agricultural products from harvesting fields to warehouses or transport vehicles, a type of agricultural materials handling robot was designed, the optimal trajectory planning method for the collaborative operating time of agricultural materials handling robots was proposed, and the time optimal trajectory under the collaborative operation of robot operating system and traveling system was acquired. Specifically, the kinematic model and dynamic model for the collaborative operation of robots were established to perform time optimal trajectory planning for materials handling robots, the Beta distribution was then applied to the Whale Optimization Algorithm (WOA) for population initialization, and a nonlinear convergence factor was introduced to prevent local optimum in the later stage of iterations. Finally, WOA was improved combining the variable neighborhood algorithm to enhance the diversity of the neighborhood structure, and this improved algorithm was applied to model solving. The results reveal that the proposed trajectory planning method can facilitate robots to obtain a smooth and time optical moving trajectory in collaborative operations of materials grabbing and discharging and obstacle avoidance. The displacement, speed, acceleration, and force/torque curves of each joint of the robots change gently, and the double-crawler traction can meet the requirements of the robots and rapidly stabilize and track the time optical trajectory.

An operator-inspired framework for metaheuristics and its applications on job-shop scheduling problems

The job-shop scheduling problem (JSP) is a well-known combinatorial optimization problem in manufacturing systems. For the past two decades, real-number metaheuristics have been widely used to solve the JSP using the real-number transform methods. A limitation of the real-number metaheuristics is the premature convergence due to the stochasticity of the transform methods. To eliminate this limitation, a novel operator framework has been designed, building the bridge between the discrete optimization problem (JSP) and real-number metaheuristics (also called continuous metaheuristics). Specifically, this paper captures the core operators of the real-number metaheuristics, namely, addition, subtraction, and multiplication. Firstly, three new operators are reconstructed according to several simple neighborhood structures to solve the JSP robustly and effectively. The properties of the arithmetic (symmetry) are taken into account in the reconstruction of the proposed operators to avoid excessive redundant searches. Secondly, a positional similarity-based population diversity is presented for the JSP to demonstrate the intrinsic distinctions between the proposed operators and the transform methods during the evolutionary process. Finally, the results of five widely used benchmark test suites (185 instances) show that the proposed operators can achieve a better balance between exploration and exploitation than the current real-number transform methods.

Expressive Multi-Agent Communication via Identity-Aware Learning

Information sharing through communication is essential for tackling complex multi-agent reinforcement learning tasks. Many existing multi-agent communication protocols can be viewed as instances of message passing graph neural networks (GNNs). However, due to the significantly limited expressive ability of the standard GNN method, the agent feature representations remain similar and indistinguishable even though the agents have different neighborhood structures. This further results in the homogenization of agent behaviors and reduces the capability to solve tasks effectively. In this paper, we propose a multi-agent communication protocol via identity-aware learning (IDEAL), which explicitly enhances the distinguishability of agent feature representations to break the diversity bottleneck. Specifically, IDEAL extends existing multi-agent communication protocols by inductively considering the agents' identities during the message passing process. To obtain expressive feature representations for a given agent, IDEAL first extracts the ego network centered around that agent and then performs multiple rounds of heterogeneous message passing, where different parameter sets are applied to the central agent and the other surrounding agents within the ego network. IDEAL fosters expressive communication between agents and generates distinguishable feature representations, which promotes action diversity and individuality emergence. Experimental results on various benchmarks demonstrate IDEAL can be flexibly integrated into various multi-agent communication methods and enhances the corresponding performance.

Threshold-Based Responsive Simulated Annealing for Directed Feedback Vertex Set Problem

As a classical NP-hard problem and the topic of the PACE 2022 competition, the directed feedback vertex set problem (DFVSP) aims to find a minimum subset of vertices such that, when vertices in the subset and all their adjacent edges are removed from the directed graph, the remainder graph is acyclic. In this paper, we propose a threshold-based responsive simulated annealing algorithm called TRSA for solving DFVSP. First, we simplify the problem instances with two new reduction rules proposed in this paper and eight reduction rules from the literature. Then, based on a new solution representation, TRSA solves DFVSP with a fast local search procedure featured by a swap-based neighborhood structure and three neighborhood acceleration strategies. Finally, all these strategies are incorporated into a threshold-based responsive simulated annealing framework. Computational experiments on 140 benchmark instances show that TRSA is highly competitive compared to the state-of-the-art methods. Specifically, TRSA can improve the best known results for 53 instances, while matching the best known results for 79 ones. Furthermore, some important features of TRSA are analyzed to identify its success factors.

Improving PTM Site Prediction by Coupling of Multi-Granularity Structure and Multi-Scale Sequence Representation

Protein post-translational modification (PTM) site prediction is a fundamental task in bioinformatics. Several computational methods have been developed to predict PTM sites. However, existing methods ignore the structure information and merely utilize protein sequences. Furthermore, designing a more fine-grained structure representation learning method is urgently needed as PTM is a biological event that occurs at the atom granularity. In this paper, we propose a PTM site prediction method by Coupling of Multi-Granularity structure and Multi-Scale sequence representation, PTM-CMGMS for brevity. Specifically, multigranularity structure-aware representation learning is designed to learn neighborhood structure representations at the amino acid, atom, and whole protein granularity from AlphaFold predicted structures, followed by utilizing contrastive learning to optimize the structure representations. Additionally, multi-scale sequence representation learning is used to extract context sequence information, and motif generated by aligning all context sequences of PTM sites assists the prediction. Extensive experiments on three datasets show that PTM-CMGMS outperforms the state-of-the-art methods. Source code can be found at https://github.com/LZY-HZAU/PTM-CMGMS.

Discovering Heterogeneous Causal Effects in Relational Data

Causal inference in relational data should account for the non-IID nature of the data and the interference phenomenon, which occurs when a unit's outcome is influenced by the treatments or outcomes of others. Existing solutions to causal inference under interference consider either homogeneous influence from peers or specific heterogeneous influence contexts (e.g., local neighborhood structure). This thesis investigates causal reasoning in relational data and the automated discovery of heterogeneous causal effects under arbitrary heterogeneous peer influence contexts and effect modification.

Numerical Evaluation of Vibration Induced by Explosive Blasting in the Bauxite Mines of Sangarédi, Guinea

During rocks blasting by explosives in mines and quarries, Vibration phenomena appear in the surrounding environment, often causing significant damage to neighboring structures such as: buildings, bridges, tunnels and dams. This is why mining companies using the technique of the explosion are often faced with constraints of limiting the vibration level in order to minimize or eliminate potential damage to neighboring structures or reduce neighbor complaints.
 This paper presents a new numerical prediction model for vibration level induced by explosive blasting in the Bauxite mines of Sangarédi, Guinea. The dynamic damage law is associated with vibration phenomena analyses to determine numerical value of the particle velocities of different points located at different distances from the explosion hole for the explosion of a single charge hole, in order to compare to the vibration level required by the specifications of the Guinea mining company, which authorizes a vibration level of 10 mm/s in an inhabited area. The prediction of the vibration level gives particle velocities of 11m/s at a distance of 10 m from the explosion hole. Based on the results of the model, one could estimate that this required vibration level would be reached at a distance of 80 m from the blast hole.

Graph convolutional networks for spatial interpolation of correlated data

Several deep learning methods for spatial data have been developed that report good performance in a big data setting. These methods typically require the choice of an appropriate kernel and some tuning of hyperparameters, which are contributing reasons for poor performance on smaller data sets.In this paper, we propose a mathematical construction of a graph-based neural network for spatial prediction that substantially generalizes the KCN model in [Appleby, Liu and Liu (2020). Kriging convolutional networks. In Proc. AAAI Conf. AI 34, pp. 3187–3194]. In particular, our model, referred to as SPONGE, allows for integrated learning of the convolutional kernel, admits higher order neighborhood structures and can make use of the distance between locations in the neighborhood and between labels of neighboring nodes. All of this yields higher flexibility in capturing spatial correlations.We investigate in simulation studies including small, medium and (reasonably) large data sets in what situations and to what extent SPONGE comes close to or (if the conditions for optimality are violated) even beats universal Kriging, whose predictions incur a high computational cost if n is large. Furthermore we study the improvement for general SPONGE in comparison with the usual KCN.Finally, we compare various graph-based neural network models on larger real world data sets and apply our method to the prediction of soil organic carbon in the southern part of Malawi.

The last-mile delivery vehicle routing problem with handling cost in the front warehouse mode

This paper aims to investigate the vehicle routing problem faced by a logistics service provider that performs the last-mile delivery of large household appliances under the front warehouse mode for multiple enterprises. We formulate the problem as a new variant of the pickup and delivery problem and present a mixed integer programming model for it. To solve this problem heuristically, the neighborhood structures with newly proposed move operators are explored within a variable neighborhood search (VNS) and a randomized tabu thresholding (RTT). Experiments on small instances clearly indicate the effectiveness of the two heuristic methods. The comparative results on large-scale instances show that RTT performs relatively better than VNS. In addition, through comparison with computational results of algorithms from the literature, RTT proves good ability to solve a related problem. Furthermore, we introduce a sample case to explore the benefit of FWM for last-mile delivery and to discuss the impact of inventory in front warehouses, handling cost, and split delivery on the solution.

Dynamic multi-label feature selection algorithm based on label importance and label correlation

Multi-label distribution is a popular direction in current machine learning research and is relevant to many practical problems. In multi-label learning, samples are usually described by high-dimensional features, many of which are redundant or invalid. This paper proposes a multi-label static feature selection algorithm to solve the problems caused by high-dimensional features of multi-label learning samples. This algorithm is based on label importance and label relevance, and improves the neighborhood rough set model. One reason for using neighborhood rough sets is that feature selection using neighborhood rough sets does not require any prior knowledge of the feature space structure. Another reason is that it does not destroy the neighborhood and order structure of the data when processing multi-label data. The method of mutual information is used to achieve the extension from single labels to multiple labels in the multi-label neighborhood; through this method, the label importance and label relevance of multi-label data are connected. In addition, in the multi-label task scenario, features may be interdependent and interrelated, and features often arrive incrementally or can be extracted continuously; we call these flow features. Traditional static feature selection algorithms do not handle flow features well. Therefore, this paper proposes a dynamic feature selection algorithm for flow features, which is based on previous static feature selection algorithms. The proposed static and dynamic algorithms have been tested on a multi-label learning task set and the experimental results show the effectiveness of both algorithms.

Balancing and scheduling of assembly line with multi-type collaborative robots

Human–robot collaboration (HRC) is a promising production mode that is in line with the vision of human-centered Industry 4.0. HRC contributes to the improvement of productivity as well as the reduction of workers’ ergonomic risk. In this study, we present one of the first attempts to address the assembly line balancing problem with multi-type collaborative robots (cobots), which allows human and robots to perform tasks in parallel or in collaboration. A mixed-integer programming (MIP) model is formulated to minimize the cycle time and a tight lower bound is proposed. We further propose a multi-objective model and an extended model to expand the scope of the study. Due to the complexity, an adaptive neighborhood simulated annealing algorithm (ANSA) is developed with the designed neighborhood operators and structures. Furthermore, an adaptive mechanism is applied to the ANSA to dynamically update the weights of the neighborhood structures based on historical information. Extensive computational experiments and a real case study are conducted to verify the superiority of ANSA. We further compare the application of diverse collaboration modes, i.e., sequential, simultaneous and supportive modes. The results also indicate that a suitable type of robot can improve productivity and the utilization rate of robots.

Unsupervised feature extraction based on uncorrelated approach

In high-dimensional spaces, mathematically driven data processing methods have recently attracted a lot of attention. We consider the situation when information is obtained by sampling a probability distribution with support on or close to a sub-manifold of Euclidean space. In this paper, we provide an innovative unsupervised learning method called Uncorrelated Neighborhood Preserving Embedding (UNPE) which identifies the underlying manifold structure of a data set and preserves the neighborhood structure of the data set. We provide a concrete formulation with UNPE, an iterative technique to demonstrate the usefulness of the framework, which has been confirmed by experimental findings on datasets Coil20, Pie, Tox, and Prostate-GE that uses three different parameters viz., F-score, NMI, and accuracy. It is observed that performance is better than the LPP algorithm by 1%, PCA by 2%, and more than 2% of LLE and LE algorithms.

Study on flexible job shop scheduling problem considering energy saving

The Flexible Job Shop Scheduling Problem (FJSP) is an extension of the classical Job Shop Scheduling Problem (JSP). The research objective of the traditional FJSP mainly considers the completion time, but ignores the energy consumption of the manufacturing system. In this paper, a mathematical model of the energy-efficient flexible job shop scheduling problem is constructed. The optimization objectives are completion time, delay time, and total equipment energy consumption. To solve the model, an improved non-dominated sorting genetic algorithm (CT-NSGA-II) is proposed to obtain the optimal scheduling solution. First, the heuristic rules of GLR were used to generate the initial population with good quality and diversity. Second, different crossover and variation operators are designed for the process sequencing and equipment selection parts to enhance the diversity of the evolutionary population. The sparsity theory is introduced to find sparse solutions and three neighborhood structures are designed to perform local search on sparse solutions to improve the uniformity of the optimal solution set distribution. Finally, a competitive selection strategy based on the bidding mechanism is proposed for the Pareto optimal solution set to obtain a better scheduling scheme. The experimental results show that the proposed improved algorithm is feasible and effective in the FJSP problem considering energy consumption, and the algorithm has some application value in improving the efficiency of smart shop operation.

Constructing modified variable neighborhood search approaches to solve a nurse scheduling problem

This study proposed multiple revised variable neighborhood search (VNS) approaches applying the greedy concept to solve a nurse scheduling problem (NSP). In this paper, we developed three greedy-neighbourhood-swapping mechanisms (greedy-2-exchange, greedy-3-exchange, and greedy-4-exchange) to conduct local searches based on one-, two-, or three-neighbourhood structures that accounted for constraints imposed by government and hospital regulations. The greedy-neighbourhood-swapping mechanisms were used to identify medical staff members with the highest soft-constraint (e.g. nurses’ preferences) violation weights on a given day who then swapped their shifts with others. To validate the proposed VNS approaches, we also conducted a case study. Based on the testing instances, all of the proposed VNS approaches generated optimal or near-optimal solutions, and the differences between them were small. The optimal number of the neighbourhood structures was determined to be two, confirming that a larger number of neighbourhoods in a neighbourhood structure would not necessarily be associated with more easily escaping local optima. Furthermore, the resulting outcomes supported the conclusion that the proposed modified VNS approaches generated better schedules for the medical staff members of hospitals than the compared meta-heuristic algorithms.

Self-paced semi-supervised feature selection with application to multi-modal Alzheimer’s disease classification

Semi-supervised multi-modal learning has attracted much attention due to the expense and scarcity of data labels, especially in disease diagnosis field. Most existing methods follow the paradigm by iteratively inferring the pseudo-labels of unlabeled data and add them into training sequence, but they ignore the reliability of those pseudo-labels, where inaccurate and wrong supervision will lead to negative influence on model learning. In this paper, we propose a Self-paced Semi-supervised Multi-modal Feature Selection (SSMFS) method, and apply it to Alzheimer’s disease classification. Specifically, SSMFS projects multi-modal biomedical data into the common label space with discriminative feature selection. Under the guidance of prior multi-modal similarity graphs, a unified graph is adaptively learned and embedded to preserve the neighborhood structures. More importantly, SSMFS dynamically investigates the discriminability and credibility of pseudo-labels, and adaptively assigns a weight to each unlabeled sample via self-paced learning such that the negative influence of wrong supervision can be reduced. Finally, a multi-kernel support vector machine is used to fuse the selected multi-modal features for final disease prediction. Experimental results on the Alzheimer’s Disease Neuroimaging Initiative (ADNI) dataset demonstrate the effectiveness of our method.