Bi-typed multi-relational heterogeneous graph (BMHG) is one of the most common graphs in practice, for example, academic networks, e-commerce user behavior graph and enterprise knowledge graph. It is a critical and challenge problem on how to learn the numerical representation for each node to characterize subtle structures. However, most previous studies treat all node relations in BMHG as the same class of relation without distinguishing the different characteristics between the intra-type relations and inter-type relations of the bi-typed nodes, causing the loss of significant structure information. To address this issue, we propose a novel D ual H ierarchical A ttention N etworks (DHAN) based on the bi-typed multi-relational heterogeneous graphs to learn comprehensive node representations with the intra-type and inter-type attention-based encoder under a hierarchical mechanism. Specifically, the former encoder aggregates information from the same type of nodes, while the latter aggregates node representations from its different types of neighbors. Moreover, to sufficiently model node multi-relational information in BMHG, we adopt a newly proposed hierarchical mechanism. By doing so, the proposed dual hierarchical attention operations enable our model to fully capture the complex structures of the bi-typed multi-relational heterogeneous graphs. Experimental results on various tasks against the state-of-the-arts sufficiently confirm the capability of DHAN in learning node representations on the BMHGs.

Temporal graph representation learning has drawn considerable attention in recent years. Most existing works mainly focus on modeling local structural dependencies of temporal graphs. However, underestimating the inherent global structural role information in many real-world temporal graphs inevitably leads to sub-optimal graph representations. To overcome this shortcoming, we propose a novel Role-based Temporal Graph Convolution Network (RTGCN) that fully leverages the global structural role information in temporal graphs. Specifically, RTGCN can effectively capture the static global structural roles by using hypergraph convolution neural networks. To capture the evolution of nodes' structural roles, we further design structural role-based gated recurrent units. Finally, we integrate structural role proximity in our objective function to preserve global structural similarity, further promoting temporal graph representation learning. Experimental results on multiple real-world datasets demonstrate that RTGCN consistently outperforms state-of-the-art temporal graph representation learning methods by significant margins in various temporal link prediction and node classification tasks. Specifically, RTGCN achieves AUC improvement of up to 5.1% for link prediction and F1 improvement of up to 6.2% for new link prediction. In addition, RTGCN achieves AUC improvement up to 4.6% for node classification and 2.7% for structural role classification.

The exceptionally high bandwidth requirements associated with the delivery of live 360° video content pose significant challenges in the current network context. An avenue for addressing this bandwidth challenge is to use the limited network resources for sending the user's Field-of-View (FoV) tiles at a high resolution, instead of transmitting all frame components at high quality. However, precisely forecasting the FoV for 360° live video content distribution remains a complex endeavor due to the lack of pre-knowledge on user viewing behaviors. In this paper, we present GL360, a novel 360° transmission framework, which employs Graph Representation Learning for FoV prediction. First, we analyze the interaction between users and tiles in panoramic videos utilizing a dynamic heterogeneous Relational Graph Convolutional Network (RGCN), which facilitates efficient user and tile embedding representation learning. Secondly, we propose an online dynamic heterogeneous graph learning (DHGL)-based algorithm to dynamically capture the time-varying features of the user's viewing behaviors with limited prior knowledge. Further, we design a FoV-aware content delivery algorithm that allows the edge servers to determine the video tiles' resolution for each accessed user. Experimental results based on real traces demonstrate how our solution outperforms four other solutions in terms of FoV prediction and network performance