Research map / Graph learning / Graph algorithms and clustering
Learning Hierarchical Graph Neural Networks for Image Clustering
Yifan Xing, Tong He, Tianjun Xiao, Yongxin Wang, Yuanjun Xiong, Wei Xia, David Wipf, Zheng Zhang, Stefano Soatto
Read the paperCheck your idea against 137,217 papers
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 |
|---|---|---|---|
| add hierarchy (Hi-LANDER) to single-level LANDER (ablation) | Hi-LANDER · Hannah | Fb: 0.234 -> 0.700 | Improved |
| add hierarchy (Hi-LANDER) to single-level LANDER (ablation) | Hi-LANDER · Hannah | NMI: 0.644 -> 0.810 | Improved |
| add hierarchy (Hi-LANDER) to single-level LANDER (ablation) | Hi-LANDER · Hannah | runtime (s): 44.9 -> 36.9 | Improved |
| use joint inference (single LANDER model) instead of two separate GNN models (ablation) | LANDER · Hannah | runtime (s): 256.2 -> 44.9 | Improved |
| use joint inference (single LANDER model) instead of two separate GNN models (ablation) | LANDER · Hannah | Fb: 0.224 -> 0.234 | Improved |
| early-stop agglomeration before full convergence (ablation) | Hi-LANDER · clustering benchmarks | Fp/Fb/NMI | No clear effect |
| train on a different cluster-size distribution than test (Hi-LANDER vs GCN-V+E) | Hi-LANDER · Hannah | Fb: 0.224 -> 0.700 | Improved |
| use Hi-LANDER pseudo labels instead of best baseline (GCN-V) | face recognition · IJBC 1:1 | FNMR@FMR=1e-4: 0.181 -> 0.159 | Improved |
| use Hi-LANDER vs GCN-V on same-distribution iNat2018 | Hi-LANDER · iNat2018-Test (same distribution) | Fb: 0.360 vs 0.350 | No clear effect |
| use Hi-LANDER instead of GCN-V on IMDB-Test-SameDist | Hi-LANDER · IMDB-Test-SameDist | Fp: 0.722 -> 0.793 | Improved |
Closest papers by meaning
- GC-Flow: A Graph-Based Flow Network for Effective Clustering (ICML 2023 · similarity 0.90)
- Non-Graph Data Clustering via $\mathcal {O}(n)$ Bipartite Graph Convolution (TPAMI 2022 · similarity 0.90)
- Do Graph Neural Networks Learn Generalizable Algorithms for Clustering Graphs? (NeurIPS 2026 · similarity 0.89)
- From Logits to Hierarchies: Hierarchical Clustering made Simple (ICML 2025 · similarity 0.88)
- Vision HGNN: An Image is More than a Graph of Nodes (ICCV 2023 · similarity 0.87)
- HiGen: Hierarchical Graph Generative Networks (ICLR 2024 · similarity 0.87)
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.
