Detecting anomalous edges and nodes in dynamic networks is critical in various areas, such as social media, computer networks, and so on. Recent approaches leverage network embedding technique to learn how to generate node representations for normal training samples and detect anomalies deviated from normal patterns. However, most existing network embedding approaches learn deterministic node representations, which are sensitive to fluctuations of the topology and attributes due to the high flexibility and stochasticity of dynamic networks. In this paper, a stochastic neural network, named by Hierarchical Variational Graph Recurrent Autoencoder (H-VGRAE), is proposed to detect anomalies in dynamic networks by the learned robust node representations in the form of random variables. H-VGRAE is a semi-supervised model to capture normal patterns in training set by maximizing the likelihood of the adjacency matrix and node attributes via variational inference. The encoder of the H-VGRAE encodes hierarchical spatial-temporal information of topology and node attribute into multi-layer conditional random variables, and then the decoder reconstructs the dynamic network based on the latent random variables. For a new observation of the dynamic network, the reconstruction probabilities of edges and node attributes can be obtained from the trained H-VGRAE, and those with low reconstruction probabilities are declared as anomalous. Comparing with existing methods, H-VGRAE has three main advantages: 1) H-VGRAE learns robust node representations through stochasticity modeling and the extraction of multi-scale spatial-temporal features; 2) H-VGRAE can be extended to deep structure with the increase of the dynamic network scale; 3) the anomalous edge and node can be located and interpreted from the probabilistic perspective. Extensive experiments on four realworld datasets demonstrate the outperformance of H-VGRAE on anomaly detection in dynamic networks compared with state-ofthe-art competitors.
Corporate social responsibility (CSR) has steadily grown in importance. We show government regulation on corporate reporting of CSR, aimed to spur its growth and increase transparency, has grown in tandem. Such reporting regulation is more readily observable than CSR itself and can be used as a proxy for the latter. We show that larger economies with higher institutional capacity find it easier to develop reporting regulations, and that international influences and local pollution increase concerns are important contributing factors. We show that such regulation also increases CSR, even after accounting for common unobserved factors that may affect both.
With the deepened exploration and development of petroleum, pre-stack inversion can predict reservoir and petroleum distribution by extracting elastic parameters by virtue of the gathers. Because of the abundant information, pre-stack inversion has gradually become a focus of research. Existing studies indicate that pre-stack amplitude variation with angle simultaneous inversion has a favourable application effect in conventional shallow sandstone gas reservoirs, but has poor impact in deeply hidden-type reservoirs. For improved the precision of description and prediction of hidden-type reservoirs, the Shahejie Formation in the Gangbei District of the Gangzhong Oil Field was used as an example, and a method of predicting complicated structure-lithology reservoir based on improved Fatti reflection coefficient approximation formula was proposed in this study. Based on petrophysical analysis and coordinate transformation theory, with elastic parameters which were obtained by the new technology, the indicative factors of reservoir sensitivity were established, resulting in forming a series of accurate reservoir prediction methods. Results show that the improved inversion method can accurately describe sand bodies in a complicated structure-lithology reservoir. Indicative factors of reservoir sensitivity show favourable linear correlation with sand bodies and it can indicate reservoir distribution more effectively than the conventional method. Prediction results are in agreement with actual drilling results. The study provides a theoretical reference for identifying complicated structure-lithology reservoirs.
In order to satisfy transmission and application, this paper illustrates the drawback of the current rate-distortion model and the characteristics of multi-view video coding (MVC). This paper proposes a bit allocation and rate control scheme for MVC based on frame complexity and human visual characteristics. Firstly, the proposed algorithm improves the quadratic rate-distortion (R-D) model. Then, the proposed algorithm reasonably allocates bit-rate among views based on frame complexity and human visual characteristics. This paper turns the bit allocation among views into a multi-objective optimization problem. Finally, this paper proposes a view layer and frame layer bit allocation algorithm for MVC according to frame complexity and the previously coded information. Simulation results show that the proposed algorithm can effectively control bit rate for MVC, where the average rate control error of our algorithm is 1.08%.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.