Research map / Graph learning / Graph self-supervised learning
[Re] CrossWalk: Fairness-enhanced Node Representation Learning
Luca Pantea, Andrei-Eusebiu Blahovici
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Use CrossWalk instead of FairWalk and Adversarial embedding on DeepWalk embeddings | k-medoids seed selection for influence maximization · Rice-Facebook and 2-group synthetic | disparity of influence | Improved |
| Use CrossWalk instead of FairWalk | DeepWalk embeddings for influence maximization · Twitter (3 groups) | disparity and total influence | Improved |
| Use CrossWalk instead of FairWalk on Node2Vec | Node2Vec embeddings · Rice-Facebook influence maximization | disparity of influence | Improved |
| Use CrossWalk instead of FairWalk with label propagation | DeepWalk embeddings · Rice-Facebook node classification | disparity of classification | Improved |
| Use CrossWalk instead of FairWalk for logistic regression on embeddings | DeepWalk embeddings · Rice-Facebook and Twitter link prediction | disparity across link types | Improved |
| Use CrossWalk instead of FairWalk | DeepWalk embeddings for link prediction · Rice-Facebook and Twitter link prediction | accuracy | Hurt |
| Use CrossWalk reweighting on networks with fewer inter-group edges (ablation) | CrossWalk · Synthetic 2-group graphs, varying P_inter | disparity of influence | Improved |
| Increase bias strength p when alpha is small (ablation) | CrossWalk · Synthetic graphs, P_inter 0.01 and 0.015 | disparity of influence | No clear effect |
| Increase bias parameter p above 2 (ablation) | CrossWalk · Synthetic 2-group graph | disparity of influence | Improved |
| Apply CrossWalk when inter-group probability is near intra-group probability (ablation) | CrossWalk · Synthetic graph, P_inter 0.015 | disparity of influence | No clear effect |
Closest papers by meaning
- Chasing Fairness in Graphs: A GNN Architecture Perspective (AAAI 2024 · similarity 0.83)
- Reproducibility Study Of Learning Fair Graph Representations Via Automated Data Augmentations (NeurIPS 2024 · similarity 0.83)
- FairGP: A Scalable and Fair Graph Transformer Using Graph Partitioning (AAAI 2025 · similarity 0.83)
- Fair Graph U-Net: A Fair Graph Learning Framework Integrating Group and Individual Awareness (AAAI 2025 · similarity 0.82)
- DANCE: Dual Unbiased Expansion with Group-acquired Alignment for Out-of-distribution Graph Fairness Learning (ICML 2025 · similarity 0.80)
- FairSIN: Achieving Fairness in Graph Neural Networks through Sensitive Information Neutralization (AAAI 2024 · similarity 0.80)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
