Research map / Graph learning / GNN architectures
Rethinking Graph Lottery Tickets: Graph Sparsity Matters
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| Add the Wasserstein-distance auxiliary loss head to guide edge pruning (ablation) | UGS-style GNN sparsification · Cora, Citeseer, and PubMed | test accuracy | Helped |
| Increase the auxiliary-loss weight from 0.1 to 1 (ablation) | Proposed GNN sparsification · High graph sparsity | test accuracy | Hurt |
| Use min-max adversarial optimization for the graph mask instead of UGS optimization (ablation) | GNN sparsification with WD loss (UGS+WD comparison) · High graph sparsity | test accuracy | Helped |
| Increase graph sparsity relative to weight sparsity (ablation) | Pruned GNN · Cora, Citeseer, and PubMed | test accuracy | Hurt |
| Transfer the post-trained source-task winning ticket rather than a randomly initialized GNN | 3-layer GCN transferred from arXiv to MAG · MAG | target-task test accuracy | Helped |
| Transfer the re-initialized source-task winning ticket rather than a randomly initialized GNN | 3-layer GCN transferred from arXiv to MAG · MAG | target-task test accuracy | Helped |
| Increase GLT sparsity (ablation) | GLT transferred from arXiv to MAG · MAG | transfer ratio | Hurt |
| Prune graph edges beyond the low-sparsity regime | UGS · Cora | test accuracy | Hurt |
| Add graph sparsification to GCN on PubMed, where pruning can remove noisy edges | GCN · PubMed | test accuracy | Helped |
Closest papers by meaning
- A Unified Lottery Ticket Hypothesis for Graph Neural Networks (ICML 2021 · similarity 0.96)
- Searching Lottery Tickets in Graph Neural Networks: A Dual Perspective (ICLR 2023 · similarity 0.94)
- Fast Track to Winning Tickets: Repowering One-Shot Pruning for Graph Neural Networks (AAAI 2025 · similarity 0.94)
- TEDDY: Trimming Edges with Degree-based Discrimination Strategy (ICLR 2024 · similarity 0.91)
- Weisfeiler and Leman Go Gambling: Why Expressive Lottery Tickets Win (ICML 2025 · similarity 0.91)
- Find A Winning Sign: Sign Is All We Need to Win the Lottery (ICLR 2025 · similarity 0.90)
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.