AHNA: Adaptive representation learning for attributed heterogeneous networks

Efficient sampling of complex network with modified random walk strategies

Representation learning of networks has witnessed significant progress in recent times. Such representations have been effectively used for classic network-based machine learning tasks like node classification, link prediction, and network alignment. However, very few methods focus on capturing representations for multiplex or multilayer networks, which are more accurate and detailed representations of complex networks. In this work, we propose Multi-Net a fast and scalable embedding technique for multiplex networks. Multi-Net, effectively maps nodes to a lower-dimensional space while preserving its neighborhood properties across all the layers. We utilize four random walk strategies in our unified network embedding model, thus making our approach more robust than existing state-of-the-art models. We demonstrate superior performance of Multi-Net on four real-world datasets from different domains. In particular, we highlight the uniqueness of Multi-Net by leveraging it for the complex task of network reconstruction.

In this paper, we explore different Markovian random walk strategies on networks with transition probabilities between nodes defined in terms of functions of the Laplacian matrix. We generalize random walk strategies with local information in the Laplacian matrix, that describes the connections of a network, to a dynamic determined by functions of this matrix. The resulting processes are non-local allowing transitions of the random walker from one node to nodes beyond its nearest neighbors. We find that only two types of Laplacian functions are admissible with distinct behaviors for long-range steps in the infinite network limit: type (i) functions generate Brownian motions, type (ii) functions Lévy flights. For this asymptotic long-range step behavior only the lowest non-vanishing order of the Laplacian function is relevant, namely first order for type (i), and fractional order for type (ii) functions.In the first part, we discuss spectral properties of the Laplacian matrix and a series of relations that are maintained by a particular type of functions that allow to define random walks on any type of undirected connected networks. Once described general properties, we explore characteristics of random walk strategies that emerge from particular cases with functions defined in terms of exponentials, logarithms and powers of the Laplacian as well as relations of these dynamics with non-local strategies like Lévy flights and fractional transport. Finally, we analyze the global capacity of these random walk strategies to explore networks like lattices and trees and different types of random and complex networks.

Recent researches on multiview learning have received widespread attention due to the increasing generalization of multiview data. As an effective probabilistic model, random walk has also shown encouraging performance in various fields. To further exploit the potential of utilizing random walk schemes to address multiview learning problems, this article proposes a simple yet efficient multigraph random walk scheme for both multiview clustering and semisupervised classification tasks. The proposed model integrates random walk with multiview learning, and recursively learns a globally stable probability distribution matrix from multiple views, on the basis of which the label indicator is obtained in the scene of clustering or semisupervised classification. Furthermore, an adaptive weight vector is learned to incorporate the diversity and complementarity of multiview data. Besides, the relationships between the proposed scheme and spectral clustering, neighborhood embedding and manifold embedding are analyzed theoretically. Finally, comprehensive comparative experiments are conducted with several state-of-the-art multiview clustering and semisupervised classification methods on eight real-world datasets. The experimental results demonstrate the superiority of the proposed method in terms of both clustering and classification performance.

We study the diffusive transport of Markovian random walks on arbitrary networks with stochastic resetting to multiple nodes. We deduce analytical expressions for the stationary occupation probability and for the mean and global first passage times. This general approach allows us to characterize the effect of resetting on the capacity of random walk strategies to reach a particular target or to explore the network. Our formalism holds for ergodic random walks and can be implemented from the spectral properties of the random walk without resetting, providing a tool to analyze the efficiency of search strategies with resetting to multiple nodes. We apply the methods developed here to the dynamics with two reset nodes and derive analytical results for normal random walks and Lévy flights on rings. We also explore the effect of resetting to multiple nodes on a comb graph, Lévy flights that visit specific locations in a continuous space, and the Google random walk strategy on regular networks.

Network embedding maps the nodes of a network to a continuous vector space, which can then be used as the input to downstream tasks, such as node classification, node clustering, link prediction, and similarity search. To learn network embedding more effectively, many technologies adopt the approach of random walk to obtain the network structure. As the meta-path of heterogeneous networks emerges, network embedding will be equipped with more semantic interpretation. Consequently, various random walks, based on meta-path strategies, have been proposed for network embedding. However, the combination of semantic and structure in a heterogeneous network cannot achieve ideal results. To overcome this challenge, we start from a task-guided issue by combining the timestamps information in the heterogeneous network, and then employing the method of temporal segmentation to decompose the network into a continuous temporal sequence. Finally, the set of context-paths between nodes is calculated in a continuous vector by the depth-first meta-path search algorithm. More precisely, we propose a Temporal Sliding Density Walk (TSDW) algorithm by combining network semantics and structure effectively. Empirical results for network data show that TSDW could significantly outperform the state-of-the-art representation learning models, including DeepWalk, LINE, Node2vec, PTE, Meapath2vec, HIN2vec, HTNE, and CTDNE by 3.02% to 44.9% of Macro-F1, 0.9% to 18.92% of Micro-F1 in multi-class node classification and 21% to 47% of NMI in node clustering.

Info2vec: An aggregative representation method in multi-layer and heterogeneous networks

In smart grid, smart meters are deployed to collect power consumption data periodically, and the data are analyzed to improve the efficiency of power transmission and distribution. The collected consumption data may leak the usage patterns of domestic appliances, so that it may damage the behavior privacy of customers. Most related work to protect data privacy in smart grid relies on cryptographic primitives, for example, encryption, which induces a large amount of power consumption overhead. In this paper, we make the first attempt to propose solutions without any cryptographic computation to protect user privacy. The privacy in smart grid is formally defined in the paper. Three schemes are proposed: random perturbation scheme (RPS), random walk scheme (RWS), and distance-bounded random walk with perturbation scheme (DBS). Three algorithms are also proposed in each scheme, respectively. All schemes are ultra-lightweight in terms of computation without relying any cryptographic primitive. The privacy, soundness, and accuracy of proposed schemes are guaranteed and justified by strict analysis.

Grey wolf optimizer (GWO) is a widely applied nature-inspired optimization algorithm with simple search mechanism, which suffers the deficiency of premature convergence. To improve its exploration performance and avoid trapping in to stagnation, we integrate the basic GWO with two random walk strategies, Gaussian random walk and Levy walk, and this new GWO extension is named RWGWO. The performance of RWGWO is compared with those of other comprising GWO variants by using several standard benchmarks. Moreover, RWGWO is applied in solving constrained engineering design problems. With a view to the experimental results, RWGWO has remarkable performance compared with the competitors in this work in terms of convergence speed and accuracy.

Balancing topology structure and node attribute in evolutionary multi-objective community detection for attributed networks

With the rapid deployment of graph neural networks (GNNs) based techniques in a wide range of applications such as link prediction, community detection, and node classification, the explainability of GNNs become an indispensable component for predictive and trustworthy decision making. To achieve this goal, some recent works focus on designing explainable GNN models such as GNNExplainer, PGExplainer, and Gem. These GNN explainers have shown remarkable performance in explaining the predictive results from GNNs. Despite their success, the robustness of these explainers is less explored in terms of vulnerabilities of GNN explainers. Graph perturbations such as adversarial attacks can lead to inaccurate explanations and consequently cause catastrophes. Thus, in this paper, we take the first step and strive to explore the robustness of GNN explainers. To be specific, we first define two adversarial attack scenarios—aggressive adversary and conservative adversary to contaminate graph structures. We then investigate the impacts of the poisoned graphs on the explainability of three prevalent GNN explainers with three standard evaluation metrics: \(Fidelity^+\), \(Fidelity^-\), and Sparsity. We conduct experiments on synthetic and real-world datasets and focus on two popular graph mining tasks: node classification and graph classification. Our empirical results suggest that GNN explainers are generally not robust to the adversarial attacks caused by graph structural noises.KeywordsGraph neural networksGNN explainersAdversarial attacksRobustness

Trajectory similarity computation is a fundamental problem in various applications (e.g., transportation optimization, behavioral study). Recent researches learn trajectory representations instead of point matching to realize more accurate and efficient trajectory similarity computation. However, these methods can still not be scaled to large datasets due to high computational cost. In this paper, we propose a novel hash learning method to encode the trajectories into binary hash codes and compute trajectory similarities by Hamming distances which is much more efficient. To the best of our knowledge, this is the first work to conduct hash learning for trajectory similarity computation. Furthermore, unlike the Word2Vec model based on random walk strategy, we utilize hypergraph neural networks for the first time to learn the representations for the grids by constructing the hyperedges according to the real-life trajectories, resulting in more representative grid embeddings. In addition, we design a residual network into the multi-layer GRU to learn more discriminative trajectory representations. The proposed <u>H</u>ypergraph <u>H</u>ash <u>L</u>earning for <u>T</u>rajectory similarity commutation is an end-to-end framework and named HHL-Traj. Experimental results on two real-world trajectory datasets (i.e., Porto and Beijing) demonstrate that the proposed framework achieves up to 6.23% and 15.42% accuracy gains compared with state-of-the-art baselines in unhashed and hashed cases, respectively. The efficiency of trajectory similarity computation based on hash codes is also verified. Our code is available at https://github.com/caoyuan57/HHL-Traj.

A random walk scheme, consisting of alternating phases of regular Brownian motion and Lévy walks, is proposed as a model for run-and-tumble bacterial motion. Within the continuous-time random walk approach we obtain the long-time and short-time behavior of the mean squared displacement of the walker as depending on the properties of the dwelling time distribution in each phase. Depending on these distributions, normal diffusion, superdiffusion, and ballistic spreading may arise.

The utilization of graph neural networks in heterogeneous networks has been rapidly increasing in recent years. The topological structure, fundamental to the nature of graphs, is often claimed to be jointly addressed with attribute information by most heterogeneous graph neural network (HGNN) approaches. However, this assertion lacks sufficient empirical validation. We empirically challenge these claims within existing evaluation frameworks. We propose a straightforward multi-layer perceptron (MLP) based model which relies solely on adjacency relationships for feature aggregation, without fully exploiting the graph's intricate topological structures. Unexpectedly, the model achieved nearly state-of-the-art performance on several real-world datasets. This discrepancy indicates that current evaluation protocols contain systemic biases and often fail to distinguish superficial adjacency exploitation from genuine structural pattern learning. It also points to the need for future benchmark datasets and metrics that more faithfully reflect the real requirements of topological learning. We hope this work stimulates further research into the role and significance of topological information in HGNNs.

Dynamic heterogeneous graph representation learning with neighborhood type modeling