Network Structure Research Articles

Flying by night: Comparing nocturnal pollinator networks over time in the Colorado Rocky Mountains

Abstract Where climate is seasonal, such as in high‐altitude ecosystems, the occurrence and activity patterns of insect pollinators may vary with the varying occurrence of floral resources. This high turnover of species occurrence and interactions may lead to changes in the structure and properties of plant–pollinator networks. Here, we systematically compared the properties of moth pollen‐transport networks in Colorado, USA, over the course of a season, and tested for a link between moth species specialisation and morphology. We predicted that network properties would change through the season as the abundance and diversity of available floral resources increased. We also predicted that there would be a link between moth specialisation and morphology, as moth proboscis length can restrict their ability to utilise different floral resources. We sampled moths for pollen over 9 weeks in June through August 2021. We then constructed pollen‐transport networks, and used mixed models to test for differences in network indices between different parts of the 2021 season; we also tested for a relationship between moth proboscis length, the duration of moth species' presence across the season and the number of flower species pollinated. As expected, we found high rates of species turnover and high rates of plant–pollinator interactions over the summer. However, linkage density and robustness were the only overall network metrics that changed. We also found that longer proboscis lengths and more persistent species occurrence over the season were predictive of higher generalisation among moth species. We suggest that even with high rates of species turnover, nocturnal moth pollen‐transport networks may maintain their overall structure over a season, with variations in the availability of floral resources driving limited variation in some network metrics. Our study highlights the necessity of continuing to research moth pollen‐transport networks over different temporal scales, to better understand how ecological factors can lead to changes in network structure over time.

Inverse Coupled Simulated Annealing for Enhanced OSPF Convergence in IoT Networks

The current Internet of Things (IoT) network structure is evolving from small-scale distributed systems to a large-scale hierarchical collaboration between backbone and access networks. In this context, the dynamic changes in backbone node connections and the surge in service demands, coupled with sluggish fault detection speeds, significantly shorten effective service transmission time. To address this issue, this paper proposes an inverse coupled simulated annealing for enhanced OSPF route convergence in IoT networks (OSPF-ICSA). Initially, the link state is derived from the statistical characteristics of Hello packets, while the aggregated characteristics of the link state are employed to characterize the node state, providing data support for the reverse coupled simulated annealing algorithm. Subsequently, the Hello packet is refined, and a mechanism is designed to synchronize OSPF intervals and transmit node states. This ensures that nodes within the same subnet synchronize their sending intervals and fault detection times while sharing their node states. Finally, building upon this foundation, the reverse coupled simulated annealing algorithm is introduced to jointly optimize the Hello packet sending interval and fault detection time. Compared to the traditional AODV protocol, OSPF-ICSA reduces the average fault detection time by over 37.38%, improves the average fault detection accuracy by more than 3.1%, decreases the average routing overhead by over 20%, and increases the average packet delivery rate by over 5.1%.

Glacier retreat decreases mutualistic network robustness over spacetime

Glaciers are retreating worldwide at an ever‐increasing rate, exposing new ice‐free areas to ecological succession. This process leads to changes in biodiversity and potentially to novel species interactions. However, we still have a limited understanding of how glacier retreat influences species interaction networks, particularly the structure and robustness of mutualistic networks. After reconstructing plant–pollinator networks along a 140‐years chronosequence on a glacier foreland, we address the effects of glacier retreat on network structure and robustness. Our results show that the prevalence of different network motifs changes over spacetime, leading to a decrease of network robustness. With glacier retreat, mutualistic networks shift from highly connected with diverse specialist interactions to loosely connected with few generalist interactions. Furthermore, despite the turnover of plant species, we find that species structural roles remain constant over spacetime while depending on species identity. Our findings suggest that glacier retreat reshuffles mutualistic networks with motifs posing low robustness, leading to increased fragility. Understanding the assembly and breaking down of species interaction networks provides novel insights into the development and stability of novel, post‐glacial ecological systems facing glacier extinction.

State analysis of a biobased hydrogel subjected to freeze and thaw processes by X ray absorption spectroscopy using synchrotron radiation

Freezing hydrogels can compromise their network structures and modify their properties as a result of ice crystal formation. Therefore, understanding the internal structure, including ice crystals and the state of chemical components within hydrogels, is essential. In this study, we evaluated the elemental distribution in bio-based hydrogels subjected to freezing-thaw process using X-ray absorption spectroscopy with synchrotron radiation. A bio-based hydrogel, prepared from Alaska pollock, underwent both slow and rapid freezing processes. Tomographic images and linear X-ray absorption coefficient distributions of the rapidly frozen hydrogel displayed a uniform image with a mean absorption coefficient of 2.81 cm−1. Conversely, the slowly frozen sample exhibited distinct contrasts with peaks at 2.516 cm−1 (dark) and 3.691 cm−1 (bright), occupying 28% and 72% of the image, respectively. The mean absorption coefficient of the slowly frozen sample was comparable to that of the rapidly frozen sample, indicating no elemental loss. The elements within the hydrogel were categorized into organic elements, macrominerals, and trace elements. The bright areas in the images were attributed to the concentration of macrominerals. Notably, Cl and Na were the primary contributors to the absorption coefficients among the elements present, signifying salt migration during freezing. These findings suggest that the contrast observed in X-ray computed tomography images after freezing reflects the elemental distribution within the hydrogel and successfully demonstrates element localization due to cryoconcentration.

Bibliometric Analysis of Spline Regression Model for Trend Mapping and Strategy Development Research Using Vosviewer

Bibliometric analysis is a quantitative method used to measure and analyze scientific literature. This technique involves the collection and analysis of publication data, such as journal articles, books, and conferences, to evaluate and understand patterns of scientific communication, research productivity, author collaboration, journal impact, and trends within a scientific field. Bibliometric analysis has a limitation in that it is challenging to visualize effectively, necessitating the use of software for proper visualization. Therefore, this article discusses bibliometric analysis on spline regression for trend mapping and strategy development using VOSviewer software. The research results show that the visualization of spline regression keywords with VOSviewer can help in understanding patterns of relationships between variables, research trends, and network structures in scientific literature. Based on the analysis results, bibliometric analysis on spline regression can be visualized in trend mapping, which can aid in planning further research strategies, including identifying collaboration opportunities and underexplored research areas.

Convolutional neural networks uncover the dynamics of human visual memory representations over time.

The ability to accurately retrieve visual details of past events is a fundamental cognitive function relevant for daily life. While a visual stimulus contains an abundance of information, only some of it is later encoded into long-term memory representations. However, an ongoing challenge has been to isolate memory representations that integrate various visual features and uncover their dynamics over time. To address this question, we leveraged a novel combination of empirical and computational frameworks based on the hierarchal structure of convolutional neural networks and their correspondence to human visual processing. This enabled to reveal the contribution of different levels of visual representations to memory strength and their dynamics over time. Visual memory strength was measured with distractors selected based on their shared similarity to the target memory along low or high layers of the convolutional neural network hierarchy. The results show that visual working memory relies similarly on low and high-level visual representations. However, already after a few minutes and on to the next day, visual memory relies more strongly on high-level visual representations. These findings suggest that visual representations transform from a distributed to a stronger high-level conceptual representation, providing novel insights into the dynamics of visual memory over time.

Medical image recognition based on multilayer neural network

As a common accidental injury, accurate identification of burns and scalded injuries is of great significance to the development of treatment programs and prognosis assessment. Traditional identification methods are subjective and inaccurate, this paper proposes a burns and scalds grade identification technique based on multilayer neural network. The three-degree classification standard of burns and scalds is described, and the structure and principle of multilayer neural network are introduced, including the network structure composed of input layer, hidden layer and output layer, forward propagation, back propagation and training process. The implementation steps of burns and scald grade recognition technology based on multi-layer neural network are discussed in detail, such as data acquisition, image pre-processing, feature extraction, classifier design, model training and evaluation. Through experiments, the model is tested using a dataset containing 362 burns and scalds images, and the model achieves more than 90% accuracy on the test set with high accuracy and reliability. The multi-layer neural network model is used to classify the burn and scald data, which improves the diagnostic accuracy, shortens the delay time of the disease, reduces the pressure of professional doctors, and helps patients have a preliminary understanding of the degree of burn.

Observing the dynamic community structure of urban travel networks based on navigation data

Dynamic community division is crucial for comprehending the interactive structure and mechanism between residents and space. Currently, there is insufficient research on dynamic community changes in travel networks regarding travel distance. This study utilizes residents’ travel trajectories to construct a multi-layer network based on travel purposes. The study establishes methods for identifying travel network communities and analyzing community evolution influence. Dynamic characteristics of travel network communities at different distances are extracted. Results show that within 3–7 km travel distance, different travel network communities exhibit continuous spatial aggregation, whereas beyond 11 km, heterogeneity is observed. Moreover, communities within 3-7 km experience drastic dynamic evolution, whereas beyond 7 km, they tend to stabilize, forming a large-scale and fixed travel mode. This study enriches the dynamic interaction research of functional space from the perspective of distance, and the experimental results provide an effective reference for urban planning and management.

Network creation with homophilic agents

AbstractNetwork Creation Games are an important framework for understanding the formation of real-world networks. These games usually assume a set of indistinguishable agents strategically buying edges at a uniform price, which leads to the formation of a network among them. However, in real life, agents are heterogeneous and their relationships often display a bias towards similar agents, say of the same ethnic group. This homophilic behavior on the agent level can then lead to the emergent global phenomenon of social segregation. We study Network Creation Games with multiple types of homophilic agents and non-uniform edge cost, introducing two models focusing on the perception of same-type and different-type neighbors, respectively. Despite their different initial conditions, both our theoretical and experimental analysis show that both the composition and segregation strength of the resulting stable networks are very similar, indicating a robust structure of social networks under homophily.

Well-connectedness and community detection

Community detection methods help reveal the meso-scale structure of complex networks. Integral to detecting communities is the expectation that communities in a network are edge-dense and “well-connected”. Surprisingly, we find that five different community detection methods–the Leiden algorithm optimizing the Constant Potts Model, the Leiden algorithm optimizing modularity, Infomap, Markov Cluster (MCL), and Iterative k-core (IKC)–identify communities that fail even a mild requirement for well-connectedness. To address this issue, we have developed the Connectivity Modifier (CM), which iteratively removes small edge cuts and re-clusters until communities are well-connected according to a user-specified criterion. We tested CM on real-world networks ranging in size from approximately 35,000 to 75,000,000 nodes. Post-processing of the output of community detection methods by CM resulted in a reduction in node coverage. Results on synthetic networks show that the CM algorithm generally maintains or improves accuracy in recovering true communities. This study underscores the importance of network clusterability–the fraction of a network that exhibits community structure–and the need for more models of community structure where networks contain nodes that are not assigned to communities. In summary, we address well-connectedness as an important aspect of clustering and present a scalable open-source tool for well-connected clusters.

Effects of simplicial centrality on robustness of higher-order multiplex world trade networks

Abstract World trade networks are exhaustively described by pairwise interactions, and overlook higher-order structure from the outcome of collective interactions at the level of groups of nodes like multilateral trade agreements. To address this limitation, \textcolor{red}{we collect multiplex world trade networks, including the bilateral regional trade agreement network, which represents pairwise interactions; the multilateral regional trade agreement network, which naturally represents a higher-order network structure; and the import and export trade network, which represents pairwise interactions and additional complexities.} The analysis of simplicial centrality, including degree, closeness, and subgraph at 0, 1, and 2-simplex \textcolor{red}{levels, reveals that intra-level correlations are high, while inter-levels may exhibit significant disparities.} Nodes with low centrality at higher-order levels could influence network robustness due to the diversity of interactions and higher-order dependencies. Simplicial centrality on robustness of multiplex world trade networks under random and targeted attacks reveals that the complex connectivity of higher-order levels renders them more vulnerable post-attack. An optimization strategy of the rebalancing of network centrality is proposed to enhance the robustness, and the simulation shows risks posed to central nodes are minimized and opportunities for peripheral nodes to partake in global trade are broadened.

Adaptive Robust Control of Nonlinear Constrained Stirred‐Tank Reactors via Self‐Organizing Fuzzy Neural Network

ABSTRACTThis paper presents an adaptive robust constraint controller for a continuous stirred tank reactor (CSTR) system based on a self‐organizing fuzzy neural network (SOFNN). Due to the high complexity of the chemical reactions, the CSTR system contains many strong nonlinearities and uncertainties. This is the first time to introduce the SOFNN into the adaptive controller design of the CSTR system, which improves the adaptability to dynamic system changes through the adjustment of the fuzzy network structure. Meanwhile, the time‐varying integral barrier Lyapunov functions (TVIBLFs) are employed to ensure the dimensionless reactant concentration and mixture temperature within a reasonable scope, which can improve the stability and safety of the CSTR system. Based on Lyapunov stability analysis, all the signals in a closed‐loop system are ultimately bounded. Simulation results substantiate the efficacy of the proposed control scheme.

Autonomous and ubiquitous in-node learning algorithms of active directed graphs and its storage behavior

The brain’s memory system is extraordinarily complex, evidenced by the multitude of neurons involved and the intricate electrochemical activities within them, as well as the complex interactions among neurons. Memory research spans various levels, from cellular and molecular to cognitive behavioral studies, each with its own focus, making it challenging to fully describe the memory mechanism. Many details of how biological neuronal networks encode, store, and retrieve information remain unknown. In this study, we model biological neuronal networks as active directed graphs, where each node is self-adaptive and relies on local information for decision-making. To explore how these networks implement memory mechanisms, we propose a parallel distributed information access algorithm based on the node scale of the active directed graph. Here, subgraphs are seen as the physical realization of the information stored in the active directed graph. Unlike traditional algorithms with global perspectives, our algorithm emphasizes global node collaboration in resource utilization through local perspectives. While it may not achieve the global optimum like a global-view algorithm, it offers superior robustness, concurrency, decentralization, and biological feasibility. We also tested network capacity, fault tolerance, and robustness, finding that the algorithm performs better in sparser network structures.

Dynamic analysis and application of network structure control in risk conduction in the industrial chain

According to control theory, a dynamical system is controllable if, with a suitable choice of inputs, it can be driven from any initial state to any desired final state within a finite time. Most dynamic characteristics of real networks are nonlinear, so achieving target control is more practical and necessary. The network’s control energy is also a problem that must be considered. Whether and how to control the complex system of the industrial chain has high theoretical and practical significance. In this study, we use the GARCH model, DCC model, and network structure control theory comprehensively to study the price fluctuation risk of the mining stock market from the perspective of the industry chain and network control dynamics and obtain interesting results. (1) Risk conduction among stocks has a prominent industry-driving effect, and the risk conduction ability of upper and middle stocks is stronger. (2) The risk regulation cost, time cost, and node number cost of the whole-industry chain are all higher than those of the two-tier chain, which indicates that the correlation complexity of the network has a positive relationship with risk control. (3) Key risk nodes play an essential role in risk control, so monitoring key stocks from the industrial chain perspective is necessary to control risks in time. This work can provide valuable suggestions for market regulators and policy-makers in terms of risk management and control.

Sliding-Induced Rehydration in Hydrogels for Restoring Lubrication and Anticreeping Capability.

Fluid exudation in cartilage under normal loading can be counteracted by a sliding-induced rehydration phenomenon, which has a hydrodynamic origin related to a wedge effect at the contact inlet. Similar to cartilage, hydrogels also exhibit tribological rehydration properties, and we mimic this phenomenon to restore hydration lubrication and overcome creeping. It occurs within a specific velocity range and is mainly dependent on the applied load and hydrogel network structures. Crucially, a certain velocity in the mixed lubrication regime can produce a hydrodynamic pressure peak at the wedge and drive the rehydration inflow to overcome the extrusion. At lower sliding velocities in the boundary lubrication regime, inflows are insufficient to counteract fluid exudation, whereas at higher velocities in the hydrodynamic lubrication regime, the inlet wedge effect would diminish. These results suggest that tribological rehydration offers a novel approach to enhancing load-bearing capacity and maintaining lubrication in the hydrogels.

Asteroid material classification based on multi-parameter constraints using artificial intelligence

Material types of asteroids provide key clues to their evolutionary history and contained resources. The Gaia mission has released extensive low-resolution spectral observation data of small Solar System bodies. However, methods for classifying asteroids based on low-resolution space-based spectra are still inadequate, and do not fully leverage the complementary features of spectra and multiple intrinsic attributes of asteroids to achieve precise material classification. Our goal is to propose a method with a higher generalization accuracy for asteroid material classification by integrating multi-source information, identifying optimal feature combinations for model inputs, and deepening the understanding of relationships among asteroid parameters. The effective asteroid photometric, physical, and orbital parameters were screened using the information gain ratio and Spearman's rank correlation coefficient. Then, artificial intelligence techniques were employed to combine asteroid spectra with the selected various parameters for six-class material classification. By comparing five machine learning models, we identified network structures with higher validation accuracy and stable generalization performance. Meanwhile, feature ablation experiments were conducted to determine the input parameter combinations suitable for different scenarios. Finally, based on the statistical results and model outputs, the constraint relationships among asteroid parameters were visualized and analyzed. The proposed AsterRF model achieved a validation accuracy of 92.2<!PCT!>, an improvement of approximately 7.8 percentage points compared to existing methods that use only spectra. V-type asteroids exhibited the highest classification accuracy, followed by A-type and D-type. X-type asteroids had the lowest precision and recall, and were easily confused with C-type. The model generally showed higher classification confidence for S-type asteroids. The top five attributes that the model focused on are the phase slope parameter (G), orbital type, albedo, H magnitude, and effective diameter. Additionally, the correlations between asteroid materials and other parameters were generally below 0.4. Incorporating optimal asteroid parameter combinations can significantly enhance classification accuracy based on spectra. A dual-channel network that processes spectra and parameter inputs separately, and employs a self-attention mechanism for feature fusion is effective in combining multi-source asteroid information. Both the statistical correlations and model performance-based importance rankings of parameters contribute to understanding the constraint relationships among asteroid attributes.

Synergistic integration of fragmented transportation networks: When do networks (not) synergize?

The transportation of information, goods, people, and other entities inevitably experiences fluctuations in demand and supply, which results in the underutilization of certain links within a transportation network while other links are fully utilized and unable to accommodate any additional demand. A promising strategy for leveraging these links is the collaborative integration of fragmented transportation networks to facilitate access between isolated nodes. We present a novel analytical framework for examining the conditions under which such network integration would enhance the overall transportation efficiency. We conducted comprehensive numerical experiments to investigate the influence of the network structure on the synergistic effects of network integration with a focus on the link survival ratio and dissimilarity between networks. The results showed that networks comprising nodes with relatively uniform importance benefit significantly from network integration whereas core–periphery networks are less likely to benefit. Based on our findings, we propose an indicator for predicting the potential for synergistic effects that is related to the growth rate of the giant component. This study contributes understanding of transportation network dynamics and offers practical insights for optimizing network integration strategies to leverage underutilized capacities and improve various types of transportation systems.

Quantifying Shape and Texture Biases for Enhancing Transfer Learning in Convolutional Neural Networks

Neural networks have inductive biases owing to the assumptions associated with the selected learning algorithm, datasets, and network structure. Specifically, convolutional neural networks (CNNs) are known for their tendency to exhibit textural biases. This bias is closely related to image classification accuracy. Aligning the model’s bias with the dataset’s bias can significantly enhance performance in transfer learning, leading to more efficient learning. This study aims to quantitatively demonstrate that increasing shape bias within the network by varying kernel sizes and dilation rates improves accuracy on shape-dominant data and enables efficient learning with less data. Furthermore, we propose a novel method for quantitatively evaluating the balance between texture bias and shape bias. This method enables efficient learning by aligning the biases of the transfer learning dataset with those of the model. Systematically adjusting these biases allows CNNs to better fit data with specific biases. Compared to the original model, an accuracy improvement of up to 9.9% was observed. Our findings underscore the critical role of bias adjustment in CNN design, contributing to developing more efficient and effective image classification models.

Flexible Krylov methods for group sparsity regularization

Abstract This paper introduces new solvers for efficiently computing solutions to large-scale inverse problems with group sparsity regularization, including both non-overlapping and overlapping groups. Group sparsity regularization refers to a type of structured sparsity regularization, where the goal is to impose additional structure in the regularization process by assigning variables to predefined groups that may represent graph or network structures. Special cases of group sparsity regularization include ℓ 1 and isotropic total variation regularization. In this work, we develop hybrid projection methods based on flexible Krylov subspaces, where we first recast the group sparsity regularization term as a sequence of 2-norm penalization terms using adaptive regularization matrices in an iterative reweighted norm fashion. Then we exploit flexible preconditioning techniques to efficiently incorporate the weight updates. The main advantages of these methods are that they are computationally efficient (leveraging the advantages of flexible methods), they are general (and therefore very easily adaptable to new regularization term choices), and they are able to select the regularization parameters automatically and adaptively (exploiting the advantages of hybrid methods). Extensions to multiple regularization terms and solution decomposition frameworks (e.g. for anomaly detection) are described, and a variety of numerical examples demonstrate both the efficiency and accuracy of the proposed approaches compared to existing solvers.

Enhanced Biodegradation of Silk Fibroin Hydrogel for Preventing Postoperative Adhesion.

An absorbable adhesion barrier is a medical device that prevents postoperative adhesion and matches its biodegradation time with the regeneration period of its target tissues, which is important for antiadhesion effects. Physical hydrogels of Bombyx mori silk fibroin (SF) proteins are degradable in vivo. However, their biodegradation time is too long to exert antiadhesion effects. To shorten the biodegradation time of the SF hydrogels, we decreased the molecular weight (MW) of the SF proteins by alkaline treatment and prepared low-MW (LMW) SF hydrogels. The hydrogels contained less β-sheet crystalline and more amorphous structures than conventional, high-MW (HMW) SF hydrogels. Because of the potential loosened SF molecular structures in the hydrogel networks, the LMW SF hydrogels showed enhanced biodegradation (i.e., shorter in vitro enzymatic biodegradation time and faster in vivo biodegradation rate) as well as a lower affinity for plasma proteins and fibroblasts, which are involved in postoperative adhesion formation. An antiadhesion test using a rat abdominal adhesion model demonstrated that the LMW SF hydrogel applied to the abraded cecum was almost completely degraded within two weeks postimplantation, with a significantly lower adhesion severity score than that in the untreated model rat group. Conversely, the HMW SF hydrogel remained between the cecum and abdominal wall, with the same adhesion severity as that of the untreated model rat group. Therefore, we concluded that the antiadhesion effects of SF hydrogels were induced by enhanced biodegradation. The results of this study indicate the potential of LMW SF hydrogels as absorbable adhesion barriers.