Self-supervised Heterogeneous Graph Neural Network with Co-contrastive Learning

Wang, Xiao; Liu, Nian; Han, Hu; Shi, Chuan

doi:10.1145/3447548.3467415

Cited by 201 publications

(57 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…• HeCo [44]: It is a recently developed heterogeneous graph neural network based on the cross-view supervised learning architecture. We generate the meta-path relation from our multi-behavior interaction graph.…”

Section: 12mentioning

confidence: 99%

“…iii) The cost of InfoNCE-based mutual information calculation is 𝑂 (𝐵 × 𝑑) and 𝑂 (𝐵 × 𝑆 × 𝑑) for the numerator and denominator (in Equation3), respectively. Here, 𝑆 is the sampling size of contrastive learning for reducing the time complexity and increasing the randomness to achieve model robustness[44]. Therefore, our multi-behavior contrastive learning paradigm takes 𝑂 (𝐾 × |R 𝑘+ | × 𝑆 × 𝑑) time per epoch.…”

mentioning

confidence: 99%

See 1 more Smart Citation

Contrastive Meta Learning with Behavior Multiplicity for Recommendation

Wei

Huang

Xia

et al. 2022

Proceedings of the Fifteenth ACM International Conference on Web Search and Data Mining

104

View full text Add to dashboard Cite

A well-informed recommendation framework could not only help users identify their interested items, but also benefit the revenue of various online platforms (e.g., e-commerce, social media). Traditional recommendation models usually assume that only a single type of interaction exists between user and item, and fail to model the multiplex user-item relationships from multi-typed user behavior data, such as page view, add-to-favourite and purchase. While some recent studies propose to capture the dependencies across different types of behaviors, two important challenges have been less explored: i) Dealing with the sparse supervision signal under target behaviors (e.g., purchase). ii) Capturing the personalized multi-behavior patterns with customized dependency modeling. To tackle the above challenges, we devise a new model CML, Contrastive Meta Learning (CML), to maintain dedicated crosstype behavior dependency for different users. In particular, we propose a multi-behavior contrastive learning framework to distill transferable knowledge across different types of behaviors via the constructed contrastive loss. In addition, to capture the diverse multi-behavior patterns, we design a contrastive meta network to encode the customized behavior heterogeneity for different users. Extensive experiments on three real-world datasets indicate that our method consistently outperforms various state-of-the-art recommendation methods. Our empirical studies further suggest that the contrastive meta learning paradigm offers great potential for capturing the behavior multiplicity in recommendation. We release our model implementation at: https://github.com/weiwei1206/CML.git. CCS CONCEPTS• Information systems → Recommender systems.

show abstract

Section: 12mentioning

confidence: 99%

mentioning

confidence: 99%

Contrastive Meta Learning with Behavior Multiplicity for Recommendation

Wei

Huang

Xia

et al. 2022

Proceedings of the Fifteenth ACM International Conference on Web Search and Data Mining

104

View full text Add to dashboard Cite

show abstract

“…DMGI [20] conduct contrastive learning on the normal graph and the corrupt graph through the meta-path. In HeCo [15], it use different views to create positive and negative pairs through metapaths and then train them in contrast. Either they did not fully utilize the characteristics of heterogeneous graphs, or they could not produce positive or negative pairs properly.…”

Section: Contrastive Learningmentioning

confidence: 99%

“…The baseline consists of three unsupervised homogeneous methods (GraphSAGE [6], GAE [26], DGI [12]) and five unsupervised heterogeneous methods ( Mp2vec [27], HERec [3], HetGNN [18], DMGI [20], HeCo [15]) and one semisupervised heterogeneous method, HAN [7].…”

Section: Baselinesmentioning

confidence: 99%

“…Since there are many non-attribute graphs in heterogeneous graphs, when augmentation of the existing attribute masking method is used, even if BGRL is used, the maximum performance is not achieved. In order to solve this problem, a new augmentation method has been proposed [15] recently, focusing on the fact that heterogeneous graphs can be viewed from two perspectives. However, these studies require two models and cannot take advantage of the BGRL.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Cross-view Self-Supervised Learning on Heterogeneous Graph Neural Network via Bootstrapping

Park¹

2022

Preprint

View full text Add to dashboard Cite

Heterogeneous graph neural networks can represent information of heterogeneous graphs with excellent ability. Recently, self-supervised learning manner is researched which learns the unique expression of a graph through a contrastive learning method. In the absence of labels, this learning methods show great potential. However, contrastive learning relies heavily on positive and negative pairs, and generating high-quality pairs from heterogeneous graphs is difficult. In this paper, in line with recent innovations in self-supervised learning called BYOL or bootstrapping, we introduce a that can generate good representations without generating large number of pairs. In addition, paying attention to the fact that heterogeneous graphs can be viewed from two perspectives, network schema and meta-path views, high-level expressions in the graphs are captured and expressed. The proposed model showed state-of-the-art performance than other methods in various real world datasets.

show abstract

Malware Detection with Limited Supervised Information via Contrastive Learning on API Call Sequences

Gao

Pan

2022

Information and Communications Security

View full text Add to dashboard Cite

Self-supervised Heterogeneous Graph Neural Network with Co-contrastive Learning

Cited by 201 publications

References 8 publications

Contrastive Meta Learning with Behavior Multiplicity for Recommendation

Contrastive Meta Learning with Behavior Multiplicity for Recommendation

Cross-view Self-Supervised Learning on Heterogeneous Graph Neural Network via Bootstrapping

Malware Detection with Limited Supervised Information via Contrastive Learning on API Call Sequences

Contact Info

Product

Resources

About