No abstract

No abstract

With the growing popularity of social network in sport, it expresses the social relationships between individuals and facilitates realistic applications, e.g., social event mining and discovery. Sport network as a specific social network has been widely studied in research and commercial fields. However, most of the existing works utilize a simplex strategy to improve certain indicators in the team and do not consider the effect of strategy adjustment based on the current situation. In this paper, we study the problem of efficient strategy mining in football social network. To address this problem, we propose a quantitative way to combine the aspects of coordination, adaptability, flexibility, and tempo into a passing network, which notably improves the timeliness and information content of the existing network. On this basis, we design a suppression function to express the impact of strategy. Then, we propose a novel passing network and group cooperation scheme based on quantified team performance to obtain the efficient strategies. At last, the experimental results show that, based on the performance of the same team, our optimized passing network has a higher winning rate in practice.

In recent years, Graph Convolutional Networks (GCNs) have achieved great success in learning from graph-structured data. With the growing tendency of graph nodes and edges, GCN training by single processor cannot meet the demand for time and memory, which led to a boom into distributed GCN training frameworks research. However, existing distributed GCN training frameworks require enormous communication costs between processors since multitudes of dependent nodes and edges information need to be collected and transmitted for GCN training from other processors. To address this issue, we propose a Graph Augmentation based Distributed GCN framework (GAD). In particular, GAD has two main components, GAD-Partition and GAD-Optimizer. We first propose a graph augmentation-based partition (GAD-Partition) that can divide original graph into augmented subgraphs to reduce communication by selecting and storing as few significant nodes of other processors as possible while guaranteeing the accuracy of the training. In addition, we further design a subgraph variance-based importance calculation formula and propose a novel weighted global consensus method, collectively referred to as GAD-Optimizer. This optimizer adaptively reduces the importance of subgraphs with large variances for the purpose of reducing the effect of extra variance introduced by GAD-Partition on distributed GCN training. Extensive experiments on four large-scale real-world datasets demonstrate that our framework significantly reduces the communication overhead (≈ 50%), improves the convergence speed (≈ 2X) of distributed GCN training, and slight gain in accuracy (≈ 0.45%) based on minimal redundancy compared to the state-of-the-art methods.

No abstract