Loss-aware Curriculum Learning for Heterogeneous Graph Neural Networks

arXiv:2402.18875v1 Announce Type: new
Abstract: Heterogeneous Graph Neural Networks (HGNNs) are a class of deep learning models designed specifically for heterogeneous graphs, which are graphs that contain different types of nodes and edges. This paper investigates the application of curriculum learning techniques to improve the performance and robustness of Heterogeneous Graph Neural Networks (GNNs). To better classify the quality of the data, we design a loss-aware training schedule, named LTS that measures the quality of every nodes of the data …

abstract application arxiv class cs.lg curriculum curriculum learning deep learning graph graph neural networks graphs loss networks neural networks nodes paper performance robustness type types