Adaptive Neighbor Research Articles

Firefly forest: A swarm iteration-free swarm intelligence clustering algorithm

The Firefly Forest algorithm is a novel bio-inspired clustering method designed to address key challenges in traditional clustering techniques, such as the need to set a fixed number of neighbors, predefine cluster numbers, and rely on computationally intensive swarm iterative processes. The algorithm begins by using an adaptive neighbor estimation, refined to filter outliers, to determine the brightness of each firefly. This brightness guides the formation of firefly trees, which are then merged into cohesive firefly forests, completing the clustering process. This approach allows the algorithm to dynamically capture both local and global patterns, eliminate the need for predefined cluster numbers, and operate with low computational complexity. Experiments involving 14 established clustering algorithms across 19 diverse datasets, using 8 evaluative metrics, demonstrate the Firefly Forest algorithm’s superior accuracy and robustness. These results highlight its potential as a powerful tool for real-world clustering applications. Our code will be open-sourced after the peer review.

A unified feature-motion consistency framework for robust image matching

Establishing reliable feature matches between a pair of images in various scenarios is a long-standing open problem in photogrammetry. Attention-based detector-free matching with coarse-to-fine architecture has been a typical pipeline to build matches, but the cross-attention module with global receptive field may compromise the structural local consistency by introducing irrelevant regions (outliers). Motion field can maintain structural local consistency under the assumption that matches for adjacent features should be spatially proximate. However, motion field can only estimate local displacements between consecutive images and struggle with long-range displacements estimation in large-scale variation scenarios without spatial correlation priors. Moreover, large-scale variations may also disrupt the geometric consistency with the application of mutual nearest neighbor criterion in patch-level matching, making it difficult to recover accurate matches. In this paper, we propose a unified feature-motion consistency framework for robust image matching (MOMA), to maintain structural consistency at both global and local granularity in scale-discrepancy scenarios. MOMA devises a motion consistency-guided dependency range strategy (MDR) in cross attention, aggregating highly relevant regions within the motion consensus-restricted neighborhood to favor true matchable regions. Meanwhile, a unified framework with hierarchical attention structure is established to couple local motion field with global feature correspondence. The motion field provides local consistency constraints in feature aggregation, while feature correspondence provides spatial context prior to improve motion field estimation. To alleviate geometric inconsistency caused by hard nearest neighbor criterion, we propose an adaptive neighbor search (soft) strategy to address scale discrepancy. Extensive experiments on three datasets demonstrate that our method outperforms solid baselines, with AUC improvements of 4.73/4.02/3.34 in two-view pose estimation task at thresholds of 5°/10°/20° on Megadepth test, and 5.94% increase of accuracy at threshold of 1px in homography task on HPatches datasets. Furthermore, in the downstream tasks such as 3D mapping, the 3D models reconstructed using our method on the self-collected SYSU UAV datasets exhibit significant improvement in structural completeness and detail richness, manifesting its high applicability in wide downstream tasks. The code is publicly available at https://github.com/BunnyanChou/MOMA.

Adaptive Neighbors Graph Learning for Large-Scale Data Clustering using Vector Quantization and Self-Regularization

In traditional adaptive neighbors graph learning (ANGL)-based clustering, the time complexity is more than O(n2), where n is the number of data points, which is not scalable for large-scale data problems in real applications. Subsequently, ANGL adds a balance regularization to its objective function to avoid the sparse over-fitting problem in the learned similarity graph matrix. Still, the regularization may leads to many weak connections between data points in different clusters. To address these problems, we propose a new fast clustering method, namely, Adaptive Neighbors Graph Learning for Large-Scale Data Clustering using Vector Quantization and Self-Regularization (ANGL-LDC), to perform vector quantization (VQ) on original data and feed the obtained VQ data as the input in the n×n similarity graph matrix learning. Hence, the n×n similarity graph matrix learning problem is simplified to weighted m×m(m≪n) graph learning problem, where m is the number of distinct points and weight is the duplicate times of distinct points in VQ data. Consequently, the time complexity of ANGL-LDC is much lower than that of ANGL. At the same time, we propose a new ANGL objective function with a graph connection self-regularization mechanism, where the ANGL-LDC objective function will get an infinity value if the value of one graph connection is equal to 1. Therefore, ANGL-LDC naturally avoids obtaining the sparse over-fitting problem since we need to minimize the value of ANGL-LDC’s objective function. Experimental results on synthetic and real-world datasets demonstrate the scalability and effectiveness of ANGL-LDC.

Graph attention networks with adaptive neighbor graph aggregation for cold-start recommendation

Graph attention networks with adaptive neighbor graph aggregation for cold-start recommendation

Online Unsupervised Multi-View Feature Selection with Adaptive Neighbors

Online Unsupervised Multi-View Feature Selection with Adaptive Neighbors

Sparse multi-view image clustering with complete similarity information

Multi-view image clustering aims to efficiently divide the collection of images by studying the characteristics of different views. Many studies performed Laplacian dimensionality reduction on the original image to avoid noise interference in constructing the similarity matrix. However, they ignored the problem that local similarity information and isolated image information are lost due to dimensionality reduction. These problems will result in the similarity matrix being insufficient to accurately describe the similarity between images, which in turn affects the clustering accuracy. Therefore, we propose a sparse multi-view spectral clustering model with complete similarity information between images (SMSC-CSI). The model combines the original image space and the low-dimensional spectral embedding space based on the adaptive neighbors method to learn the initial similarity matrices of each view jointly. On the one hand, the original space can retain all the similarity information between images so that the initial similarity matrix can accurately describe the similarity between images, which is conducive to accurate clustering. On the other hand, the low-dimensional space can avoid noise interference and retain the main structure of high-dimensional data, improving the robustness of the initial similarity matrix. Meanwhile, to ensure consistency among the views, the model minimizes the difference between the initial similarity matrix and the central fusion matrix by alternately updating to obtain the optimal weights and central fusion matrix for each view. Finally, we can obtain ideal clustering results directly with low model complexity and without post-processing steps such as K-means by adding non-negative Laplace rank constraints and ℓ0-norm constraints to the model objective function. Experimental results on different real image datasets show that SMSC-CSI can outperform some traditional clustering models and recent models on multi-view image clustering.

Probabilistic Nearest Neighbors Based Locality Preserving Projections for Unsupervised Metric Learning

Dimensionality reduction based unsupervised metric learning consists in finding meaningful compact data representations previously to clustering and classification problems. One of the major aspects of these algorithms is the approximation of the underlying manifold by a weighted graph. A limitation with most manifold learning algorithms is that edge weights in the proximity graph rely heavily on the Euclidean distance, which is known to be quite sensitive to the presence of outliers. In this paper, we propose to improve the Locality Preserving Projections (LPP) algorithm by incorporating a recently proposed graph inference method called Probabilistic Nearest Neighbors (PNN), an extension of the Clustering with Adaptive Neighbors (CAN) approach, used with success in graph-based semi-supervised learning. The proposed PNN-LPP algorithm is able to achieve better classification results than regular LPP, showing competitive performance against state-of-the-art approaches for dimensionality reduction, such as the UMAP algorithm, especially in datasets with a limited number of samples.

Algorithms For Cold-Start Game Recommendation Based On GNN Pre-training Model

Abstract In the absence of sufficient user behavior data, game recommendation systems face the cold-start problem. To address this issue, this paper proposes a solution based on the Graph Neural Network pre-training model to alleviate the cold-start problem. The proposed model directly reconstructs cold-start user/game embeddings using a meta-learning setup based on dataset training simulations and uses an adaptive neighbor sampler to improve user interaction relations and thereby to improve game recommendation performance. Experimental results demonstrate the effectiveness and practicality of the recommendation model proposed in this study. Moreover, the proposed model is embedded in the game recommendation system to visualize the recommendation results.

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.

Consensus Affinity Graph Learning via Structure Graph Fusion and Block Diagonal Representation for Multiview Clustering

Learning a robust affinity graph is fundamental to graph-based clustering methods. However, some existing affinity graph learning methods have encountered the following problems. First, the constructed affinity graphs cannot capture the intrinsic structure of data well. Second, when fusing all view-specific affinity graphs, most of them obtain a fusion graph by simply taking the average of multiple views, or directly learning a common graph from multiple views, without considering the discriminative property among diverse views. Third, the fusion graph does not maintain an explicit cluster structure. To alleviate these problems, the adaptive neighbor graph learning approach and the data self-expression approach are first integrated into a structure graph fusion framework to obtain a view-specific structure affinity graph to capture the local and global structures of data. Then, all the structural affinity graphs are weighted dynamically into a consensus affinity graph, which not only effectively incorporates the complementary affinity structure of important views but also has the capability of preserving the consensus affinity structure that is shared by all views. Finally, a k–block diagonal regularizer is introduced for the consensus affinity graph to encourage it to have an explicit cluster structure. An efficient optimization algorithm is developed to tackle the resultant optimization problem. Extensive experiments on benchmark datasets validate the superiority of the proposed method.

Optimizing multimodal timetable synchronization of intercity railway and metro for the first service period during holidays

Metro plays a vital role in managing passenger distribution at intercity railway (IR) stations, particularly during holidays when there is a surge in tourist traffic. To efficiently accommodate the high demand for intercity travel, it becomes imperative for metro agencies to optimize holiday timetables. This paper focuses on designing holiday timetables of the first service period for the metro network that connects to an IR station, aiming to enhance multimodal collaboration with IR timetables while ensuring seamless coordination among various metro lines at the network level. A bi-objective model is proposed to maximize the temporal availability of metro network and minimize transfer waiting times for IR passengers traveling in early morning. To solve the model, an improved Artificial Bee Colony algorithm is designed, incorporating adaptive neighbour search and simulated annealing techniques. The effectiveness of the model and algorithm is verified using the Shanghai Metro network and Hongqiao Railway Station. Results indicate a 9.46% increase in the temporal availability of metro network for IR passengers, coupled with a 9.68% reduction in passenger transfer waiting times. Notably, the study reveals that solely advancing operations of the IR-connected metro lines is inefficient. Instead, optimizing train timetables for the entire metro network proves to be a cost-effective approach to enhancing the overall service level of early-morning operations. Furthermore, the study emphasizes the significance of even-numbered train headways in reducing passenger transfer waiting times.

Adaptive HELLO protocol for vehicular networks

In vehicular networks, the update of car Firmware Over The Air (FOTA) is becoming a challenging issue and it mainly relies on topology discovery of neighbouring nodes. Topology discovery in mobile wireless networks is usually done by using HELLO messages. Due to mobility, topology changes occur frequently and must be quickly discovered to avoid routing failures. Since the optimal HELLO frequency depends on parameters that are subject to changes (e.g., speed of nodes, density of nodes), it must be dynamically adjusted to obtain the best trade-off between the network load and the freshness of routing tables. Existing solutions assume random mobility, constant node density and average speed, which do not hold in vehicular networks because vehicles follow specific trajectory patterns (the roads) and density and speed evolve as a function of time (rush hour vs non-rush hour) and area (urban, rural, highway). In this paper, we first draw the specific features of a vehicular network at different times and spaces by analysing real datasets and then propose a dynamic neighbour discovery protocol, Vehicular Adaptive Neighbour discovery Protocol (VANP). VANP is a fully-distributed protocol that sends beacons at an optimal frequency without knowing it a priori. The objective is to reduce the frequency at which HELLO messages are sent to save bandwidth and energy while still preserving the quality of the neighbour discovery. Through extensive simulations run on real datasets, we show that the optimal HELLO frequency can be reached by maintaining a constant optimal turnover, independent from the speed of the nodes and by aiming at this turnover, nodes automatically use the optimal HELLO frequency. Results show that VANP allows the discovery of relevant neighbours by missing at most two neighbours over all scenarios and reducing the number of HELLO messages up to twice, hence saving bandwidth and energy.

Incomplete Multiview Nonnegative Representation Learning With Graph Completion and Adaptive Neighbors.

Despite incomplete multiview clustering (IMC) being widely studied in the past decade, it is still difficult to model the correlation among multiple views due to the absence of partial views. Most existing works for IMC only mine the correlation among multiple views from available views and ignore the importance of missing views. To address this issue, we propose a novel Incomplete Multiview Nonnegative representation learning model with Graph completion and Adaptive neighbors (IMNGA), which performs common graph learning, missing graph completion, and consensus nonnegative representation learning simultaneously. In IMNGA, the common graph on all views and the incomplete graph of each view are used to reconstruct the completed graph of the corresponding view, where the common graph satisfies the neighbor constraints of incomplete multiview data and consensus representation. IMNGA gets consensus representation by factorizing completed and incomplete graphs, where consensus representation satisfies the common graph constraint. IMNGA shows its effectiveness by outperforming other state-of-the-art methods.

Joint structure bipartite graph projection and its application for industrial process monitoring

Process monitoring plays a crucial role in maintaining the safety and stable operation of industrial processes. One of its key tasks is to improve the accuracy of detecting abnormal conditions. To this end, we propose a dimensionality reduction method called Joint Structure Bipartite Graph Projection. This method not only captures global information, but also incorporates manifold learning and anchor strategy to capture the local structure of data in an adaptive neighbor manner. In constructing the bipartite graph between projected samples and projected anchors, structural information between samples and anchors is fully considered to ensure an effective graph. Furthermore, we introduce an information entropy-based method to compute the balance parameters, reducing costs of parameter tuning. Ultimately, we develop a process monitoring model based on JSBGP method, achieving the goals of anomaly detection and abnormal variables isolation. Through experiments on benchmark dataset and actual industrial process demonstrate the effectiveness of the proposed method.

Traffic Flow Prediction Based on Interactive Dynamic Spatio-Temporal Graph Convolution with a Probabilistic Sparse Attention Mechanism

Accurate traffic flow prediction is of great practical significance to alleviate road congestion. Existing methods ignore the hidden dynamic associations between road nodes, and for the problem of difficulty capturing the dynamic spatio-temporal features of traffic flow in the prediction process, a novel model based on the interactive dynamic spatio-temporal graph convolutional probabilistic sparse attention mechanism (IDG-PSAtt) is proposed, which consists of an interactive dynamic graph convolutional network (IDGCN) structure with a spatio-temporal convolutional block (ST-Conv block) and a probabilistic sparse self-attention mechanism block (ProbSSAtt block). Among them, the IDGCN synchronizes the dynamic spatio-temporal features captured by interaction sharing, and the ST-Conv block is combined with the ProbSSAtt block to effectively capture the long short-term temporal features of the traffic flow. In addition, to effectively find the hidden dynamic associations between road network nodes, a dynamic graph convolutional network generated by the fusion of an adaptive neighbor matrix and a learnable neighbor matrix was constructed. Experimental results demonstrate that the prediction performance of the IDG-PSAtt model outperforms the baseline model under the evaluation criteria and experimental settings given in this paper. In the PEMS-BAY dataset, the mean absolute error and root mean square error of the IDG-PSAtt to 60 min are improved by 15.49% and 12.10%, compared with the state-of-the-art model, respectively.

Unifying heterogeneous and homogeneous relations for personalized compatibility modeling

In recent years, the fashion industry has developed rapidly, leading to an increasing demand for personalized clothing matching. Although noteworthy progress has been made by existing work with advanced graph neural networks, there still remain three prominent challenges: comprehensive entity relation modeling, adaptive neighbor node sampling, and thorough characteri-zation of fashion items based on multi-modalities. To solve the above problems, we propose a novel personalized compatibility modeling framework, which explores both heterogeneous (i.e., item-user and user-item) and homogeneous (i.e., item-item and user-user) relation graphs with multi-modalities. Moreover, we design a correlation-guided neighbor sampling strategy to learn the correlation-aware subgraphs, which plays a crucial role in facilitating the exploration of compatibility relationships or personalized preferences. Finally, we conduct extensive experiments on two publicly available fashion datasets to demonstrate the superiority of our proposed model over existing benchmark methods.

Multi-scale topology and position feature learning and relationship-aware graph reasoning for prediction of drug-related microbes.

The human microbiome may impact the effectiveness of drugs by modulating their activities and toxicities. Predicting candidate microbes for drugs can facilitate the exploration of the therapeutic effects of drugs. Most recent methods concentrate on constructing of the prediction models based on graph reasoning. They fail to sufficiently exploit the topology and position information, the heterogeneity of multiple types of nodes and connections, and the long-distance correlations among nodes in microbe-drug heterogeneous graph. We propose a new microbe-drug association prediction model, NGMDA, to encode the position and topological features of microbe (drug) nodes, and fuse the different types of features from neighbors and the whole heterogeneous graph. First, we formulate the position and topology features of microbe (drug) nodes by t-step random walks, and the features reveal the topological neighborhoods at multiple scales and the position of each node. Second, as the features of nodes are high-dimensional and sparse, we designed an embedding enhancement strategy based on supervised fully connected autoencoders to form the embeddings with representative features and the more discriminative node distributions. Third, we propose an adaptive neighbor feature fusion module, which fuses features of neighbors by the constructed position- and topology-sensitive heterogeneous graph neural networks. A novel self-attention mechanism is developed to estimate the importance of the position and topology of each neighbor to a target node. Finally, a heterogeneous graph feature fusion module is constructed to learn the long-distance correlations among the nodes in the whole heterogeneous graph by a relationship-aware graph transformer. Relationship-aware graph transformer contains the strategy for encoding the connection relationship types among the nodes, which is helpful for integrating the diverse semantics of these connections. The extensive comparison experimental results demonstrate NGMDA's superior performance over five state-of-the-art prediction methods. The ablation experiment shows the contributions of the multi-scale topology and position feature learning, the embedding enhancement strategy, the neighbor feature fusion, and the heterogeneous graph feature fusion. Case studies over three drugs further indicate that NGMDA has ability in discovering the potential drug-related microbes. Source codes and Supplementary Material are available at https://github.com/pingxuan-hlju/NGMDA.

An efficient solution space exploring and descent method for packing equal spheres in a sphere

The problem of packing equal spheres in a spherical container is a classic global optimization problem, which has attracted enormous studies in academia and found various applications in industry. This problem is computationally very challenging, and many efforts focus on small-scale instances with the number of spherical items less than 200 in the literature. In this work, we propose an efficient local search heuristic algorithm named solution space exploring and descent for solving this problem, which can quantify the solution’s quality to determine the number of exploring actions and quickly discover a high-quality configuration. Besides, we propose an adaptive neighbor item maintenance method to speed up the convergence of the continuous optimization process and reduce the time consumption. Computational experiments on a large number of benchmark instances with 5≤n≤400 spherical items show that our algorithm significantly outperforms the state-of-the-art algorithm. Specifically, our algorithm improves 274 best-known results and matches 84 best-known results out of the 396 well-known benchmark instances.

AFSC: An Improved Spectral Clustering Based on Adaptive Neighbor and Fuzzy Affiliation

AFSC: An Improved Spectral Clustering Based on Adaptive Neighbor and Fuzzy Affiliation

Fuzzy C-means clustering algorithm based on adaptive neighbors information

Fuzzy C-means clustering algorithm based on adaptive neighbors information