Inferring the topology of a network from network dynamics is a significant problem with both theoretical research significance and practical value. This paper considers how to reconstruct the network topology according to the continuous-time data on the network. Inspired by the generative adversarial network(GAN), we design a deep learning framework based on network continuous-time data. The framework predicts the edge connection probability between network nodes by learning the correlation between network node state vectors. To verify the accuracy and adaptability of our method, we conducted extensive experiments on scale-free networks and small-world networks at different network scales using three different dynamics: heat diffusion dynamics, mutualistic interaction dynamics, and gene regulation dynamics. Experimental results show that our method significantly outperforms the other five traditional correlation indices, which demonstrates that our method can reconstruct the topology of different scale networks well under different network dynamics.

Unmanned Aerial Vehicles (UAVs) are widely used in applications such as land assessment, surveillance, and rescue operations, where they are often required to cover multiple disjoint regions. Coverage Path Planning (CPP) aims to determine optimal paths for UAVs to cover these areas. While CPP for single regions has been extensively studied, multi-region CPP with multiple UAVs remains underexplored. Existing methods typically focus on minimizing path length, but often neglect the nonlinear variations in energy consumption during flight, limiting their practical applicability. This paper addresses the multi-UAV, multi-region CPP as a variant of the Vehicle Routing Problem (VRP) with energy estimation. We propose an approach that optimizes UAV flight speeds to minimize energy consumption, supported by an accurate energy estimation algorithm. In addition, a heuristic algorithm is developed to balance the distribution of tasks among UAVs, considering both the scanning and transit times. Experiments using real-world data from the Changsha urban area demonstrate that our approach outperforms state-of-the-art methods in computational efficiency and energy savings, highlighting its potential for practical UAV deployment.

Regarding attack scenarios as query graphs and conducting subgraph matching on the data system is an important approach to identify and detect cyber threats. However, existing subgraph matching methods are not suitable for detecting time-evolving attacks since they either focus on single-query graphs or ignore the temporal constraints between multiple queries. In this paper, we model the time-evolving attack detection as a novel temporal multi-query subgraph matching problem and propose an efficient algorithm to address this problem. We first give a compact representation of the temporal query graph by merging all queries into one. Based on the temporal query graph, we propose a concise auxiliary data structure to maintain partial solutions. In addition, we employ a query matching tree to generate an efficient matching order and enumerate matchings based on the order. Extensive experiments over real-world datasets confirm the effectiveness and efficiency of our approach.

Dynamics always exist in complex systems. Graphs (complex networks) are a mathematical form for describing a complex system abstractly. Dynamics can be learned efficiently from the structure and dynamics state of a graph. Learning the dynamics in graphs plays an important role in predicting and controlling complex systems. Most of the methods for learning dynamics in graphs run slowly in large graphs. The complexity of the large graph’s structure and its nonlinear dynamics aggravate this problem. To overcome these difficulties, we propose a general framework with two novel methods in this paper, the Dynamics-METIS (D-METIS) and the Partitioned Graph Neural Dynamics Learner (PGNDL). The general framework combines D-METIS and PGNDL to perform tasks for large graphs. D-METIS is a new algorithm that can partition a large graph into multiple subgraphs. D-METIS innovatively considers the dynamic changes in the graph. PGNDL is a new parallel model that consists of ordinary differential equation systems and graph neural networks (GNNs). It can quickly learn the dynamics of subgraphs in parallel. In this framework, D-METIS provides PGNDL with partitioned subgraphs, and PGNDL can solve the tasks of interpolation and extrapolation prediction. We exhibit the universality and superiority of our framework on four kinds of graphs with three kinds of dynamics through an experiment.