Thi Linh Hoang

Thi Linh Hoang

Publications

Dynamic-GTN: Learning an Node Efficient Embedding in Dynamic Graph with Transformer

Published in Pacific Rim International Conference on Artificial Intelligence (PRICAI), 2022

In this paper, we propose the Dynamic-GTN model which is designed to learn the node embedding in a continous-time dynamic graph. The Dynamic-GTN extends the attention mechanism in a standard GTN to include temporal information of recent node interactions. Based on temporal patterns interaction between nodes, the Dynamic-GTN employs an node sampling step to reduce the number of attention operations in the dynamic graph. We evaluate our model on three benchmark datasets for learning node embedding in dynamic graphs.

Recommended citation: T. L. Hoang, and V. C. Ta, "Dynamic-GTN: Learning an Node Efficient Embedding in Dynamic Graph with Transformer", PRICAI 2022.

Effect of Cluster-based Sampling on the Over-smoothing Issue in Graph Neural Network

Published in 14th International Conference on Knowledge and Systems Engineering (KSE), 2021

In this paper, we propose the usage of cluster-based sampling to reduce the smoothing effect of the high number of layers in GNN. Given each nodes is assigned to a specific region of the embedding space, the cluster-based sampling is expected to propagate this information to the node’s neighbour, thus improve the nodes’ expressivity.

Recommended citation: T. L. Hoang, and V. C. Ta, "Effect of Cluster-based Sampling on the Over-smoothing Issue in Graph Neural Network", KSE 2022.

Improving Graph Convolutional Networks with Transformer Layer in social-based items recommendation

Published in 13th International Conference on Knowledge and Systems Engineering (KSE), 2021

In this paper, we improve the embedding output of the graph-based convolution layer by adding a number of transformer layers. The transformer layers with attention architecture help discover frequent patterns in the embedding space which increase the predictive power of the model in the downstream tasks.

Recommended citation: T. L. Hoang, T. D. Pham and V. C. Ta, "Improving Graph Convolutional Networks with Transformer Layer in social-based items recommendation", KSE 2021.