Showing papers on "Complex network published in 2020"
TL;DR: This learning method–called e-prop–approaches the performance of backpropagation through time (BPTT), the best-known method for training recurrent neural networks in machine learning and suggests a method for powerful on-chip learning in energy-efficient spike-based hardware for artificial intelligence.
Abstract: Recurrently connected networks of spiking neurons underlie the astounding information processing capabilities of the brain. Yet in spite of extensive research, how they can learn through synaptic plasticity to carry out complex network computations remains unclear. We argue that two pieces of this puzzle were provided by experimental data from neuroscience. A mathematical result tells us how these pieces need to be combined to enable biologically plausible online network learning through gradient descent, in particular deep reinforcement learning. This learning method–called e-prop–approaches the performance of backpropagation through time (BPTT), the best-known method for training recurrent neural networks in machine learning. In addition, it suggests a method for powerful on-chip learning in energy-efficient spike-based hardware for artificial intelligence. Bellec et al. present a mathematically founded approximation for gradient descent training of recurrent neural networks without backwards propagation in time. This enables biologically plausible training of spike-based neural network models with working memory and supports on-chip training of neuromorphic hardware.
281 citations
TL;DR: Li et al. as discussed by the authors proposed a lightweight pyramid networt (LPNet) for single image deraining, which adopted recursive and residual network structures to build the proposed LPNet, which has less than 8k parameters while still achieving the state-of-the-art performance on rain removal.
Abstract: Existing deep convolutional neural networks (CNNs) have found major success in image deraining, but at the expense of an enormous number of parameters. This limits their potential applications, e.g., in mobile devices. In this paper, we propose a lightweight pyramid networt (LPNet) for single-image deraining. Instead of designing a complex network structure, we use domain-specific knowledge to simplify the learning process. In particular, we find that by introducing the mature Gaussian–Laplacian image pyramid decomposition technology to the neural network, the learning problem at each pyramid level is greatly simplified and can be handled by a relatively shallow network with few parameters. We adopt recursive and residual network structures to build the proposed LPNet, which has less than 8K parameters while still achieving the state-of-the-art performance on rain removal. We also discuss the potential value of LPNet for other low- and high-level vision tasks.
221 citations
TL;DR: A deep reinforcement learning framework that can be trained on small networks to understand the organizing principles of complex networked systems, which enables us to design more robust networks against both attacks and failures.
Abstract: Finding an optimal set of nodes, called key players, whose activation (or removal) would maximally enhance (or degrade) certain network functionality, is a fundamental class of problems in network science1,2. Potential applications include network immunization3, epidemic control4, drug design5, and viral marketing6. Due to their general NP-hard nature, those problems typically cannot be solved by exact algorithms with polynomial time complexity7. Many approximate and heuristic strategies have been proposed to deal with specific application scenarios1,2,8-12. Yet, we still lack a unified framework to efficiently solve this class of problems. Here we introduce a deep reinforcement learning framework FINDER, which can be trained purely on small synthetic networks generated by toy models and then applied to a wide spectrum of influencer finding problems. Extensive experiments under various problem settings demonstrate that FINDER significantly outperforms existing methods in terms of solution quality. Moreover, it is several orders of magnitude faster than existing methods for large networks. The presented framework opens up a new direction of using deep learning techniques to understand the organizing principle of complex networks, which enables us to design more robust networks against both attacks and failures.
140 citations
TL;DR: A novel and powerful graph K-means framework, which is composed of three coupled phases in each discrete-time period, which uses a fast heuristic approach to identify those opinion leaders who have relatively high local reputation and employs a robust opinion dynamics model to simulate the evolution of the opinion matrix.
Abstract: With the explosion of social media networks, many modern applications are concerning about people's connections, which leads to the so-called social computing . An elusive question is to study how opinion communities form and evolve in real-world networks with great individual diversity and complex human connections. In this scenario, the classic K-means technique and its extended versions could not be directly applied, as they largely ignore the relationship among interactive objects. On the other side, traditional community detection approaches in statistical physics would be neither adequate nor fair: they only consider the network topological structure but ignore the heterogeneous-objects’ attributive information. To this end, we attempt to model a realistic social media network as a discrete-time dynamical system, where the opinion matrix and the community structure could mutually affect each other. In this paper, community detection in social media networks is naturally formulated as a multi-objective optimization problem (MOOP), i.e., finding a set of densely connected components with similar opinion vectors. We propose a novel and powerful graph K-means framework, which is composed of three coupled phases in each discrete-time period. Specifically, the first phase uses a fast heuristic approach to identify those opinion leaders who have relatively high local reputation; the second phase adopts a novel dynamic game model to find the locally Pareto-optimal community structure; and the final phase employs a robust opinion dynamics model to simulate the evolution of the opinion matrix. We conduct a series of comprehensive experiments on real-world benchmark networks to validate the performance of GK-means through comparisons with the state-of-the-art graph clustering technologies.
122 citations
Hangzhou Normal University1, Southern University of Science and Technology2, Shanghai University3, Tel Aviv University4, Leidos5, University of Fribourg6, Cleveland Clinic Lerner Research Institute7, Cleveland Clinic Lerner College of Medicine8, Case Western Reserve University9, Zhejiang University10
TL;DR: This review summarizes the recent developments of computational network biology, first introducing various types of biological networks and network structural properties, and then reviewing the network-based approaches, ranging from some network metrics to the complicated machine-learning methods.
121 citations
TL;DR: A novel memory interconnection Lyapunov–Krasovskii functional is structured by taking full advantage of more information of sampling interval and state, and developing some new terms to investigate the finite-time (FT) H∞ synchronization issue for complex networks with stochastic cyber attacks and random memory information exchanges.
106 citations
TL;DR: Through intensive stochastic analysis, sufficient conditions are obtained to guarantee the desired security performance for the PNBSEs, based on which the estimator gains are acquired by solving certain matrix inequalities with nonlinear constraints.
Abstract: In this paper, the partial-nodes-based state estimators (PNBSEs) are designed for a class of uncertain complex networks subject to finite-distributed delays, stochastic disturbances, as well as randomly occurring deception attacks (RODAs). In consideration of the likely unavailability of the output signals in harsh environments from certain network nodes, only partial measurements are utilized to accomplish the state estimation task for the addressed complex network with norm-bounded uncertainties in both the network parameters and the inner couplings. The RODAs are taken into account to reflect the compromised data transmissions in cyber security. We aim to derive the gain parameters of the estimators such that the overall estimation error dynamics satisfies the specified security constraint in the simultaneous presence of stochastic disturbances and deception signals. Through intensive stochastic analysis, sufficient conditions are obtained to guarantee the desired security performance for the PNBSEs, based on which the estimator gains are acquired by solving certain matrix inequalities with nonlinear constraints. A simulation study is carried out to testify the security performance of the presented state estimation method.
96 citations
TL;DR: In this paper, the authors studied the evolution of cooperation on temporal networks and found that temporality enhances cooperation, despite the fact that bursty interaction patterns generally impede cooperation and proposed a measure to quantify the amount of temporality in a network, revealing an intermediate level that maximally boosts cooperation.
Abstract: Population structure is a key determinant in fostering cooperation among naturally self-interested individuals in microbial populations, social insect groups, and human societies. Traditional research has focused on static structures, and yet most real interactions are finite in duration and changing in time, forming a temporal network. This raises the question of whether cooperation can emerge and persist despite an intrinsically fragmented population structure. Here we develop a framework to study the evolution of cooperation on temporal networks. Surprisingly, we find that network temporality actually enhances the evolution of cooperation relative to comparable static networks, despite the fact that bursty interaction patterns generally impede cooperation. We resolve this tension by proposing a measure to quantify the amount of temporality in a network, revealing an intermediate level that maximally boosts cooperation. Our results open a new avenue for investigating the evolution of cooperation and other emergent behaviours in more realistic structured populations. Population structure enables emergence of cooperation among individuals, but the impact of the dynamic nature of real interaction networks is not understood. Here, the authors study the evolution of cooperation on temporal networks and find that temporality enhances the evolution of cooperation.
94 citations
20 Apr 2020
TL;DR: A novel reinforcement learning method for Node Injection Poisoning Attacks (NIPA), to sequentially modify the labels and links of the injected nodes, without changing the connectivity between existing nodes, is proposed.
Abstract: Graph Neural Networks (GNN) offer the powerful approach to node classification in complex networks across many domains including social media, E-commerce, and FinTech. However, recent studies show that GNNs are vulnerable to attacks aimed at adversely impacting their node classification performance. Existing studies of adversarial attacks on GNN focus primarily on manipulating the connectivity between existing nodes, a task that requires greater effort on the part of the attacker in real-world applications. In contrast, it is much more expedient on the part of the attacker to inject adversarial nodes, e.g., fake profiles with forged links, into existing graphs so as to reduce the performance of the GNN in classifying existing nodes. Hence, we consider a novel form of node injection poisoning attacks on graph data. We model the key steps of a node injection attack, e.g., establishing links between the injected adversarial nodes and other nodes, choosing the label of an injected node, etc. by a Markov Decision Process. We propose a novel reinforcement learning method for Node Injection Poisoning Attacks (NIPA), to sequentially modify the labels and links of the injected nodes, without changing the connectivity between existing nodes. Specifically, we introduce a hierarchical Q-learning network to manipulate the labels of the adversarial nodes and their links with other nodes in the graph, and design an appropriate reward function to guide the reinforcement learning agent to reduce the node classification performance of GNN. The results of the experiments show that NIPA is consistently more effective than the baseline node injection attack methods for poisoning graph data on three benchmark datasets.
90 citations
TL;DR: Experimental results on synthetic and real-world networks demonstrate the superiority of the proposed algorithm over several state-of-the-art community detection algorithms for large-scale networks, in terms of both computational efficiency and detection performance.
Abstract: Evolutionary algorithms have been demonstrated to be very competitive in the community detection for complex networks. They, however, show poor scalability to large-scale networks due to the exponential increase of search space. In this paper, we suggest a network reduction-based multiobjective evolutionary algorithm for community detection in large-scale networks, where the size of the networks is recursively reduced as the evolution proceeds. In each reduction of the network, the local communities found by the elite individuals in the population are identified as nodes of the reduced network for further evolution, thereby considerably reducing the search space. A local community repairing strategy is also suggested to correct the misidentified nodes after each network reduction during the evolution. Experimental results on synthetic and real-world networks demonstrate the superiority of the proposed algorithm over several state-of-the-art community detection algorithms for large-scale networks, in terms of both computational efficiency and detection performance.
90 citations
TL;DR: This research effort investigates the relationship between network characteristics and supply chain resilience and demonstrates that utilizing a reduced list of characteristics yields performance equal to that when using a complete set of characteristics.
TL;DR: Sufficient conditions are proposed to guarantee the existence of addressed set-membership filter in terms of certain recursive matrix inequality and an optimisation problem is proposed to minimise the hyper-ellipsoid (containing the true network state) in the sense of trace.
Abstract: This paper investigates the set-membership filtering problem for a class of nonlinear time-varying complex networks with uniformly quantised measurements over redundant channels. The network output...
TL;DR: The time-evolving interconnection among murine neurons optimizes the network information flow, network robustness, and self-organization degree and has complex implications for modeling neuronal cultures and potentially on how to design biological inspired artificial intelligence.
Abstract: Understanding the mechanisms by which neurons create or suppress connections to enable communication in brain-derived neuronal cultures can inform how learning, cognition and creative behavior emerge. While prior studies have shown that neuronal cultures possess self-organizing criticality properties, we further demonstrate that in vitro brain-derived neuronal cultures exhibit a self-optimization phenomenon. More precisely, we analyze the multiscale neural growth data obtained from label-free quantitative microscopic imaging experiments and reconstruct the in vitro neuronal culture networks (microscale) and neuronal culture cluster networks (mesoscale). We investigate the structure and evolution of neuronal culture networks and neuronal culture cluster networks by estimating the importance of each network node and their information flow. By analyzing the degree-, closeness-, and betweenness-centrality, the node-to-node degree distribution (informing on neuronal interconnection phenomena), the clustering coefficient/transitivity (assessing the “small-world” properties), and the multifractal spectrum, we demonstrate that murine neurons exhibit self-optimizing behavior over time with topological characteristics distinct from existing complex network models. The time-evolving interconnection among murine neurons optimizes the network information flow, network robustness, and self-organization degree. These findings have complex implications for modeling neuronal cultures and potentially on how to design biological inspired artificial intelligence.
TL;DR: The proposed EnRenew algorithm measures the importance of nodes based on information entropy and selects a group of important nodes through dynamic update strategy and shed light on new method of node mining in complex networks for information spreading and epidemic prevention.
Abstract: Identifying a set of influential nodes is an important topic in complex networks which plays a crucial role in many applications, such as market advertising, rumor controlling, and predicting valuable scientific publications. In regard to this, researchers have developed algorithms from simple degree methods to all kinds of sophisticated approaches. However, a more robust and practical algorithm is required for the task. In this paper, we propose the EnRenew algorithm aimed to identify a set of influential nodes via information entropy. Firstly, the information entropy of each node is calculated as initial spreading ability. Then, select the node with the largest information entropy and renovate its l-length reachable nodes' spreading ability by an attenuation factor, repeat this process until specific number of influential nodes are selected. Compared with the best state-of-the-art benchmark methods, the performance of proposed algorithm improved by 21.1%, 7.0%, 30.0%, 5.0%, 2.5%, and 9.0% in final affected scale on CEnew, Email, Hamster, Router, Condmat, and Amazon network, respectively, under the Susceptible-Infected-Recovered (SIR) simulation model. The proposed algorithm measures the importance of nodes based on information entropy and selects a group of important nodes through dynamic update strategy. The impressive results on the SIR simulation model shed light on new method of node mining in complex networks for information spreading and epidemic prevention.
TL;DR: In this article, a method for identifying influencers in complex networks via the local information dimensionality is proposed, which considers the local structural properties around the central node and reduces the computational complexity.
TL;DR: Experiments show that the proposed community detection algorithm based on influential nodes (LGIEM) is able to detect communities efficiently, and achieves better performance compared to other recent methods.
TL;DR: This paper proposes a novel framework of complex network classifier (CNC) by integrating network embedding and convolutional neural network to tackle the problem of network classification and shows CNC can not only classify networks in a high accuracy and robustness, it can also extract the features of the networks automatically.
TL;DR: This work first model the Ethereum transaction records as a complex network by incorporating time and amount features of the transactions, and then design several flexible temporal walk strategies for random-walk based graph representation of this large-scale network.
Abstract: As the largest public blockchain-based platform supporting smart contracts, Ethereum has accumulated a large number of user transaction records since its debut in 2014. Analysis of Ethereum transaction records, however, is still relatively unexplored till now. Modeling the transaction records as a static simple graph, existing methods are unable to accurately characterize the temporal and multiplex features of the edges. In this brief, we first model the Ethereum transaction records as a complex network by incorporating time and amount features of the transactions, and then design several flexible temporal walk strategies for random-walk based graph representation of this large-scale network. Experiments of temporal link prediction on real Ethereum data demonstrate that temporal information and multiplicity characteristic of edges are indispensable for accurate modeling and understanding of Ethereum transaction networks.
TL;DR: In this article, the authors show that individual algorithms exhibit a broad diversity of prediction errors, such that no one predictor or family is best, or worst, across all realistic inputs, and exploit this diversity using network-based metalearning to construct a series of stacked models that combine predictors into a single algorithm.
Abstract: Most real-world networks are incompletely observed. Algorithms that can accurately predict which links are missing can dramatically speed up network data collection and improve network model validation. Many algorithms now exist for predicting missing links, given a partially observed network, but it has remained unknown whether a single best predictor exists, how link predictability varies across methods and networks from different domains, and how close to optimality current methods are. We answer these questions by systematically evaluating 203 individual link predictor algorithms, representing three popular families of methods, applied to a large corpus of 550 structurally diverse networks from six scientific domains. We first show that individual algorithms exhibit a broad diversity of prediction errors, such that no one predictor or family is best, or worst, across all realistic inputs. We then exploit this diversity using network-based metalearning to construct a series of "stacked" models that combine predictors into a single algorithm. Applied to a broad range of synthetic networks, for which we may analytically calculate optimal performance, these stacked models achieve optimal or nearly optimal levels of accuracy. Applied to real-world networks, stacked models are superior, but their accuracy varies strongly by domain, suggesting that link prediction may be fundamentally easier in social networks than in biological or technological networks. These results indicate that the state of the art for link prediction comes from combining individual algorithms, which can achieve nearly optimal predictions. We close with a brief discussion of limitations and opportunities for further improvements.
TL;DR: An innovative experimental network-based quality assessment was proposed to validate the method of identifying the importance of nodes and a generalized mechanical model is proposed that uses global information and local information.
Abstract: How to assess the importance of nodes in the network is an open question. There are many ways to identify the importance of nodes in complex networks. However, these methods have their own shortcomings and advantages. In particular, some methods based on the importance of nodes between interactions between nodes have been proposed. These methods utilize local information or path information. How to combine local and global information is still a problem. In this paper, a generalized mechanical model is proposed that uses global information and local information. To verify the effectiveness of the method, some experiments were performed on a total of ten real networks. In particular, an innovative experimental network-based quality assessment was proposed to validate the method of identifying the importance of nodes.
TL;DR: Some new analytical tools, including the method of contradiction, L’Hopital rule, and Barbalat lemma are developed to establish adaptive synchronization criteria of the addressed networks to realize asymptotical synchronization.
Abstract: In this paper, spatial diffusions are introduced to fractional-order coupled networks and the problem of synchronization is investigated for fractional-order coupled neural networks with reaction-diffusion terms. First, a new fractional-order inequality is established based on the Caputo partial fractional derivative. To realize asymptotical synchronization, two types of adaptive coupling weights are considered, namely: 1) coupling weights only related to time and 2) coupling weights dependent on both time and space. For each type of coupling weights, based on local information of the node’s dynamics, an edge-based fractional-order adaptive law and an edge-based fractional-order pinning adaptive scheme are proposed. Furthermore, some new analytical tools, including the method of contradiction, L’Hopital rule, and Barbalat lemma are developed to establish adaptive synchronization criteria of the addressed networks. Finally, an example with numerical simulations is provided to illustrate the validity and effectiveness of the theoretical results.
TL;DR: This article deals with the recursive filtering issue for a class of nonlinear complex networks (CNs) with switching topologies, random sensor failures and dynamic event-triggered mechanisms with a certain guaranteed upper bound on the filtering error covariance.
Abstract: This article deals with the recursive filtering issue for a class of nonlinear complex networks (CNs) with switching topologies, random sensor failures and dynamic event-triggered mechanisms. A Markov chain is utilized to characterize the switching behavior of the network topology. The phenomenon of sensor failures occurs in a random way governed by a set of stochastic variables obeying certain probability distributions. In order to save communication cost, a dynamic event-triggered transmission protocol is introduced into the transmission channel from the sensors to the recursive filters. The objective of the addressed problem is to design a set of dynamic event-triggered filters for the underlying CN with a certain guaranteed upper bound (on the filtering error covariance) that is then locally minimized. By employing the induction method, an upper bound is first obtained on the filtering error covariance and subsequently minimized by properly designing the filter parameters. Finally, a simulation example is provided to demonstrate the effectiveness of the proposed filtering scheme.
TL;DR: This work proposes a comprehensive framework for understanding how the interactions between genes, proteins and metabolites contribute to the determinants of robustness in a heterogeneous biological network, and designs a simulated perturbation process to characterize each gene to the overall system’s robustness.
Abstract: Robustness is a prominent feature of most biological systems. Most previous related studies have been focused on homogeneous molecular networks. Here we propose a comprehensive framework for understanding how the interactions between genes, proteins and metabolites contribute to the determinants of robustness in a heterogeneous biological network. We integrate heterogeneous sources of data to construct a multilayer interaction network composed of a gene regulatory layer, a protein–protein interaction layer, and a metabolic layer. We design a simulated perturbation process to characterize the contribution of each gene to the overall system’s robustness, and find that influential genes are enriched in essential and cancer genes. We show that the proposed mechanism predicts a higher vulnerability of the metabolic layer to perturbations applied to genes associated with metabolic diseases. Furthermore, we find that the real network is comparably or more robust than expected in multiple random realizations. Finally, we analytically derive the expected robustness of multilayer biological networks starting from the degree distributions within and between layers. These results provide insights into the non-trivial dynamics occurring in the cell after a genetic perturbation is applied, confirming the importance of including the coupling between different layers of interaction in models of complex biological systems. Robustness is a prominent feature of most biological systems, but most of the current efforts have been focused on studying homogeneous molecular networks. Here the authors propose a comprehensive framework for understanding how the interactions between genes, proteins, and metabolites contribute to the determinants of robustness.
TL;DR: By introducing the stability theory of fractional-order differential systems and the framework of Filippov regularization, some sufficient conditions are derived for ascertaining the asymptotic and finite-time cluster synchronization of coupled fractiona-order neural networks, respectively.
Abstract: This article is devoted to the cluster synchronization issue of coupled fractional-order neural networks. By introducing the stability theory of fractional-order differential systems and the framework of Filippov regularization, some sufficient conditions are derived for ascertaining the asymptotic and finite-time cluster synchronization of coupled fractional-order neural networks, respectively. In addition, the upper bound of the settling time for finite-time cluster synchronization is estimated. Compared with the existing works, the results herein are applicable for fractional-order systems, which could be regarded as an extension of integer-order ones. A numerical example with different cases is presented to illustrate the validity of theoretical results.
TL;DR: The spontaneous occurrence of synchronization phenomena that closely resemble the ones seen during epileptic seizures in humans are reported, indicating that a topology with some balance between regularity and randomness favors the self-initiation and self-termination of episodes of seizure-like strong synchronization.
Abstract: We study patterns of partial synchronization in a network of FitzHugh-Nagumo oscillators with empirical structural connectivity measured in human subjects. We report the spontaneous occurrence of synchronization phenomena that closely resemble the ones seen during epileptic seizures in humans. In order to obtain deeper insights into the interplay between dynamics and network topology, we perform long-term simulations of oscillatory dynamics on different paradigmatic network structures: random networks, regular nonlocally coupled ring networks, ring networks with fractal connectivities, and small-world networks with various rewiring probability. Among these networks, a small-world network with intermediate rewiring probability best mimics the findings achieved with the simulations using the empirical structural connectivity. For the other network topologies, either no spontaneously occurring epileptic-seizure-related synchronization phenomena can be observed in the simulated dynamics, or the overall degree of synchronization remains high throughout the simulation. This indicates that a topology with some balance between regularity and randomness favors the self-initiation and self-termination of episodes of seizure-like strong synchronization.
TL;DR: Inspired by the concept of graph convolutional networks, a simply yet effectively method named R C N N is presented to identify critical nodes with the best spreading ability and shows that under Susceptible–Infected–Recovered (SIR) model, this method outperforms the traditional benchmark methods.
Abstract: Critical nodes of complex networks play a crucial role in effective information spreading. There are many methods have been proposed to identify critical nodes in complex networks, ranging from centralities of nodes to diffusion-based processes. Most of them try to find what kind of structure will make the node more influential. In this paper, inspired by the concept of graph convolutional networks(GCNs), we convert the critical node identification problem in complex networks into a regression problem. Considering adjacency matrices of networks and convolutional neural networks(CNNs), a simply yet effectively method named R C N N is presented to identify critical nodes with the best spreading ability. In this approach, we can generate feature matrix for each node and use a convolutional neural network to train and predict the influence of nodes. Experimental results on nine synthetic and fifteen real networks show that under Susceptible–Infected–Recovered (SIR) model, R C N N outperforms the traditional benchmark methods on identifying critical nodes under spreading dynamic.
TL;DR: The proposed method starts from the payoff of the more general evolutionary game phenomena in reality, and eliminates the nodes with negative payoff and the edges connected with the failed nodes, and shows that the aggregation and invulnerability coefficients of the scale-free network are on the rise.
TL;DR: A novel sufficient criterion is derived for fixed-time synchronized of complex networks and a new finite-time stability theorem is proposed, which means the designed controller is simpler than the existing controllers by using finite-/fixed-time methods, and the chattering phenomenon can also be avoided.
Abstract: This brief investigates the fixed-time synchronization of complex networks with a simpler nonchattering controller. First, a new finite-time stability theorem is proposed. Second, a novel sufficient criterion is derived for fixed-time synchronized of complex networks. Compared with some existing results, the new controller without sign function is designed to realize fixed-time synchronization. Moreover, the designed controller without include the linear parts, which means the controller is simpler than the existing controllers by using finite-/fixed-time methods, and the chattering phenomenon can also be avoided. Finally, the application of fixed-time non-chattering controller is discussed by designing synchronization circuit of complex systems.
TL;DR: The modified CNNBCN model with a modified activation function for the magnetic resonance imaging classification of brain tumors achieves satisfactory results in brain tumor image classification and enriches the methodology of neural network design.
Abstract: The diagnosis of brain tumor types generally depends on the clinical experience of doctors, and computer-assisted diagnosis improves the accuracy of diagnosing tumor types. Therefore, a convolutional neural network based on complex networks (CNNBCN) with a modified activation function for the magnetic resonance imaging classification of brain tumors is presented. The network structure is not manually designed and optimized, but is generated by randomly generated graph algorithms. These randomly generated graphs are mapped into a computable neural network by a network generator. The accuracy of the modified CNNBCN model for brain tumor classification reaches 95.49%, which is higher than several models presented by other works. In addition, the test loss of brain tumor classification of the modified CNNBCN model is lower than those of the ResNet, DenseNet and MobileNet models in the experiments. The modified CNNBCN model not only achieves satisfactory results in brain tumor image classification, but also enriches the methodology of neural network design.
TL;DR: A novel method to identify influential nodes is proposed, which takes not only the importance of itself but also the influence of all nodes in the graph into consideration and provides a quantitative model to measure the global importance of each node (GIN).
Abstract: How to identify influential nodes in complex networks is an open issue. Several centrality measures have been proposed to address this. But these studies concentrate only on only one aspect. To solve this problem, a novel method to identify influential nodes is proposed, which takes into account not only the importance of itself but also the influence of all nodes in the graph into consideration. This approach has superiority in identifying nodes that seem unimportant but are important in the complex network. Besides, it provides a quantitative model to measure the global importance of each node (GIN). The comparison experiments conducted on six different networks illustrate the effectiveness of the proposed method.