Simulating virtual crowds can bring significant economic benefits to various applications, such as film and television special effects, evacuation planning, and rescue operations. However, the key challenge in crowd simulation is ensuring efficient and reliable autonomous navigation for numerous agents within virtual environments. In recent years, deep reinforcement learning has been used to model agents’ steering strategies, including marching and obstacle avoidance. However, most studies have focused on simple, homogeneous scenarios (e.g., intersections, corridors with basic obstacles), making it difficult to generalize the results to more complex settings. In this study, we introduce a new crowd simulation approach that combines deep reinforcement learning with anisotropic fields. This method gives agents global prior knowledge of the high complexity of their environment, allowing them to achieve impressive motion navigation results in complex scenarios without the need to repeatedly compute global path information. Additionally, we propose a novel parameterized method for constructing crowd simulation environments and evaluating simulation performance. Through evaluations across three different scenario levels, our proposed method exhibits significantly enhanced efficiency and efficacy compared to the latest methodologies. Our code is available at https://github.com/tomblack2014/DRL_Crowd_Simulation.

As a typical spatiotemporal series prediction task, traffic flow prediction has found wide application in intelligent transportation systems (ITS). Despite some progress, several unresolved issues persist. Many existing works calculate the dependencies between nodes based on stable long-term traffic data. However, the short-term dependencies are dynamically changing over time, and neglecting them would cause a decrease in predictive performance. In this article, we propose a novel dynamic graph convolution and spatiotemporal self-attention (DGSTA) network for traffic flow prediction. Specifically, considering the large amount of short-term and the dynamic dependencies between nodes, we design a new dynamic graph convolution module, which generates adjacency matrices for each time step in a day to dynamically capture the changing short-term dependencies. Additionally, we utilize a multihead spatiotemporal self-attention module to, respectively, extract static spatial and temporal correlations between nodes. Furthermore, we design a sequential embedding to explicitly model the long-term correlation between nodes. Extensive experiments conducted on three real-world datasets demonstrate that DGSTA exhibits high competitiveness. The code and data are available at https://github.com/lzmmm30/DGSTA.

Social recommendation (SocialRS), which utilizes user social information to improve recommendation performance, has received increasing attention. Graph neural networks (GNNs) facilitate the integration of both user preference and social features in SocialRS. However, existing techniques face two challenges: 1) the inherent sparse supervision signals and noise issues in real-world social networks; 2) current social recommendation methods suffer from the neglect of user preference and social attribute heterogeneity, which hinders the extraction of preference-related information from social networks. Taking inspiration from social enhancement and contrastive learning methods, we propose a social recommendation model DSVC based on dual social view contrastive learning. Specifically, in response to the first challenge, our model derives the consistency factors of users in different augmented social views, which are used to highlight noise-resistant users and jettison preference-independent social relationships in social views. To address the second challenge, we adopt probability vectors generated from consistency factors. These vectors guide the cross-view augmentation process of the interaction graph, which helps supplement social self-supervised signals and effectively avoid noise retained due to indiscriminate augmentation. The baseline model comparison experiment, ablation experiment, parameter adjustment experiment and robustness experiment conducted on three different real-world datasets consistently validated the effectiveness of our model in improving recommendation performance.