Adversarial Graph Disentanglement

Zheng, Shuai

doi:10.48550/arxiv.2103.07295

Cited by 2 publications

(3 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Current multi-intent modeling usually adopts a disentangled representation learning paradigm, which splits the user embedding into multiple chunks and tries to learn each chunked emebedding respectively for representing each user intent. In graph representation learning area, DisenGCN [13], IPGDN [11] and ADGCN [36] adopt such a paradigm and utilize the Hilbert-Schmidt Independence Criterion (HSIC) and adversarial learning for ensuring the effectiveness of intent modeling. As for recommendation scenarios, DGCF [24] proposes to disentagnle the user representations and adopts a mutual information restraint for the independence of all intents.…”

Section: Multi-intent Modelingmentioning

confidence: 99%

Knowledge Enhanced Multi-intent Transformer Network for Recommendation

Zou,

Wei,

Zhu

et al. 2024

Companion Proceedings of the ACM Web Conference 2024

View full text Add to dashboard Cite

Incorporating Knowledge Graphs (KGs) into Recommendation has attracted growing attention in industry, due to the great potential of KG in providing abundant supplementary information and interpretability for the underlying models. However, simply integrating KG into recommendation usually brings in negative feedback in industry, mainly due to the ignorance of the following two factors: i) users' multiple intents, which involve diverse nodes in KG. For example, in e-commerce scenarios, users may exhibit preferences for specific styles, brands, or colors. ii) knowledge noise, which is a prevalent issue in Knowledge Enhanced Recommendation (KGR) and even more severe in industry scenarios. The irrelevant knowledge properties of items may result in inferior model performance compared to approaches that do not incorporate knowledge. To tackle these challenges, we propose a novel approach named Knowledge Enhanced Multi-intent Transformer Network for Recommendation (KGTN), which comprises two primary modules: Global Intents Modeling with Graph Transformer, and Knowledge Contrastive Denoising under Intents. Specifically, Global Intents with Graph Transformer focuses on capturing learnable user intents, by incorporating global signals from user-itemrelation-entity interactions with a well-designed graph transformer, * Work done during internship at Taotian Group.

show abstract

Section: Multi-intent Modelingmentioning

confidence: 99%

Knowledge Enhanced Multi-intent Transformer Network for Recommendation

Zou,

Wei,

Zhu

et al. 2024

Companion Proceedings of the ACM Web Conference 2024

View full text Add to dashboard Cite

show abstract

“…4.3.2 Disentanglement. SAD already captures informative component representation corresponding to different factors, named micro-disentanglement [52]. To fully disentangle the sensitive attribute into an independent component, macro-disentanglement [52], emphasizing independence between different components, need to be considered in SAD.…”

Section: Optimization Objectivesmentioning

confidence: 99%

“…SAD already captures informative component representation corresponding to different factors, named micro-disentanglement [52]. To fully disentangle the sensitive attribute into an independent component, macro-disentanglement [52], emphasizing independence between different components, need to be considered in SAD. Thus, we employ distance correlation as a regularizer and a channel discriminator 𝑓 𝑑 as a supervisor to guide the training of FairSAD.…”

Section: Optimization Objectivesmentioning

confidence: 99%

Fair Graph Representation Learning via Sensitive Attribute Disentanglement

Zhu,

Li,

Zheng

et al. 2024

Proceedings of the ACM Web Conference 2024

View full text Add to dashboard Cite

Group fairness for Graph Neural Networks (GNNs), which emphasizes algorithmic decisions neither favoring nor harming certain groups defined by sensitive attributes (e.g., race and gender), has gained considerable attention. In particular, the objective of group fairness is to ensure that the decisions made by GNNs are independent of the sensitive attribute. To achieve this objective, most existing approaches involve eliminating sensitive attribute information in node representations or algorithmic decisions. However, such ways may also eliminate task-related information due to its inherent correlation with the sensitive attribute, leading to a sacrifice in utility. In this work, we focus on improving the fairness of GNNs while preserving task-related information and propose a fair GNN framework named FairSAD. Instead of eliminating sensitive attribute information, FairSAD enhances the fairness of GNNs via Sensitive Attribute Disentanglement (SAD), which separates the sensitive attribute-related information into an independent component to mitigate its impact. Additionally, FairSAD utilizes a channel masking mechanism to adaptively identify the sensitive attribute-related component and subsequently decorrelates it. Overall, FairSAD minimizes the impact of the sensitive attribute on GNN outcomes rather than eliminating sensitive attributes, thereby preserving task-related information associated with the sensitive attribute. Furthermore, experiments conducted on several real-world datasets demonstrate that FairSAD outperforms other state-of-the-art methods by a significant margin in terms of both fairness and utility performance. Our source code is available at https://github.com/ZzoomD/FairSAD. CCS CONCEPTS• Computing methodologies → Machine learning.

show abstract

Adversarial Graph Disentanglement

Cited by 2 publications

References 16 publications

Knowledge Enhanced Multi-intent Transformer Network for Recommendation

Knowledge Enhanced Multi-intent Transformer Network for Recommendation

Fair Graph Representation Learning via Sensitive Attribute Disentanglement

Contact Info

Product

Resources

About