Simple and Efficient Heterogeneous Graph Neural Network

Yang, Xueping; Yan, Mingyu; Pan, Shirui; Ye, Xiaochun; Fan, Dongrui

doi:10.48550/arxiv.2207.02547

Cited by 5 publications

(3 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Subsequently, the class prototype (i.e., class centroid) embeddings [38] are generated in the target node domain and the semantic domain by Eq. ( 9) and (10), respectively:…”

Section: Prototype-based Class Constraintmentioning

confidence: 99%

Semantic-aware Node Synthesis for Imbalanced Heterogeneous Information Networks

Gao¹,

Zhang²,

Chen³

et al. 2023

Preprint

View full text Add to dashboard Cite

Heterogeneous graph neural networks (HGNNs) have exhibited exceptional efficacy in modeling the complex heterogeneity in heterogeneous information networks (HINs). The critical advantage of HGNNs is their ability to handle diverse node and edge types in HINs by extracting and utilizing the abundant semantic information for effective representation learning. However, as a widespread phenomenon in many realworld scenarios, the class-imbalance distribution in HINs creates a performance bottleneck for existing HGNNs. Apart from the quantity imbalance of nodes, another more crucial and distinctive challenge in HINs is semantic imbalance. Minority classes in HINs often lack diverse and sufficient neighbor nodes, resulting in biased and incomplete semantic information. This semantic imbalance further compounds the difficulty of accurately classifying minority nodes, leading to the performance degradation of HGNNs. To tackle the imbalance of minority classes and supplement their inadequate semantics, we present the first method for the semantic imbalance problem in imbalanced HINs named Semantic-aware Node Synthesis (SNS). By assessing the influence on minority classes, SNS adaptively selects the heterogeneous neighbor nodes and augments the network with synthetic nodes while preserving the minority semantics. In addition, we introduce two regularization approaches for HGNNs that constrain the representation of synthetic nodes from both semantic and class perspectives to effectively suppress the potential noises from synthetic nodes, facilitating more expressive embeddings for classification. The comprehensive experimental study demonstrates that SNS consistently outperforms existing methods by a large margin in different benchmark datasets.

show abstract

“…Subsequently, the class prototype (i.e., class centroid) embeddings [38] are generated in the target node domain and the semantic domain by Eq. ( 9) and (10), respectively:…”

Section: Prototype-based Class Constraintmentioning

confidence: 99%

Semantic-aware Node Synthesis for Imbalanced Heterogeneous Information Networks

Gao¹,

Zhang²,

Chen³

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…In this work, we focused on homogeneous graphs. However, in reality graphs are often heterogeneous with multiple node and edge types [4]. Adaptations are necessary on both the neural and the symbolic side to apply KeGNN to heterogeneous graphs.…”

Section: Limitationsmentioning

confidence: 99%

“…In the field of deep learning, research on graph neural networks (GNNs) has gained momentum. Numerous models have been proposed for various graph topologies and applications [4] [5] [6] [7]. The key strength of GNNs is to find meaningful representations of noisy data, that can be used for various prediction tasks [8].…”

Section: Introductionmentioning

confidence: 99%

Knowledge Enhanced Graph Neural Networks

Werner¹,

Layaïda²,

Genevès³

et al. 2023

Preprint

View full text Add to dashboard Cite

Graph data is omnipresent and has a large variety of applications such as natural science, social networks or semantic web. Though rich in information, graphs are often noisy and incomplete. Therefore, graph completion tasks such as node classification or link prediction have gained attention. On the one hand, neural methods such as graph neural networks have proven to be robust tools for learning rich representations of noisy graphs. On the other hand, symbolic methods enable exact reasoning on graphs. We propose KeGNN, a neuro-symbolic framework for learning on graph data that combines both paradigms and allows for the integration of prior knowledge into a graph neural network model. In essence, KeGNN consists of a graph neural network as a base on which knowledge enhancement layers are stacked with the objective of refining predictions with respect to prior knowledge. We instantiate KeGNN in conjunction with two standard graph neural networks: Graph Convolutional Networks and Graph Attention Networks, and evaluate KeGNN on multiple benchmark datasets for node classification.

show abstract

Comparison of Machine Learning Approaches for Prediction of the Equivalent Alkane Carbon Number for Microemulsions Based on Molecular Properties

Furth,

Imel,

Zawodzinski

2024

J. Phys. Chem. A

View full text Add to dashboard Cite

The chemical properties of oils are vital in the design of microemulsion systems. The hydrophilic−lipophilic difference equation used to predict microemulsions' phase behavior expresses the oils' physiochemical properties as the equivalent alkane carbon number (EACN). The experimental determination of EACN requires knowledge of the temperature dependence of the microemulsion system and the effects of different surfactant concentrations. Thus, the experimental determination is timeintensive and tedious, requiring days to months for proper separations. Furthermore, the experiments require high purity of chemicals because microemulsions are sensitive to impurities. Our work focuses on the quick and reliable predictions of the EACN with machine learning (ML) models. Due to the immaturity of ML chemical predictions, we compare three graph neural networks (GNNs) and a gradient-boosted tree algorithm, known as XGBoost. The GNNs use the molecular structures represented as simplified molecular-input line-entry system (SMILES) codes for the initial input, which allows us to assess whether geometry optimization is necessary for reliable results. The XGBoost model also begins with the SMILES representations of the molecules but uses molecular descriptors instead of geometry optimizations. The best model tested (crystal graph convolutional neural network with Merck molecular force field-94) has an error of 1.15 EACN units of the true EACN for unknown data with the errors skewed toward zero and an R 2 score of 0.9.

show abstract

Simple and Efficient Heterogeneous Graph Neural Network

Cited by 5 publications

References 30 publications

Semantic-aware Node Synthesis for Imbalanced Heterogeneous Information Networks

Semantic-aware Node Synthesis for Imbalanced Heterogeneous Information Networks

Knowledge Enhanced Graph Neural Networks

Comparison of Machine Learning Approaches for Prediction of the Equivalent Alkane Carbon Number for Microemulsions Based on Molecular Properties

Contact Info

Product

Resources

About