Research map / Label-efficient and robust learning / Recognition and long-tail
Hierarchically Clustered Representation Learning
Su-Jin Shin, Kyungwoo Song, Il-Chul Moon
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 |
|---|---|---|---|
| Use data-specific rather than globally shared level proportions (HCRL2 vs HCRL1) (ablation) | HCRL · CIFAR-100, RCV1_v2, and 20Newsgroups | micro-averaged F-score: CIFAR-100: 0.199_{\pm 0.016} -> 0.225_{\pm 0.014}; RCV1_v2: 0.437_{\pm 0.029} -> 0.455_{\pm 0.030}; 20Newsgroups: 0.566_ | Improved |
| Jointly optimize hierarchical clustering and representation learning in the embedding space | HCRL2 · CIFAR-100, RCV1_v2, and 20Newsgroups | micro-averaged F-score: 0.201_{\pm 0.008} -> 0.225_{\pm 0.014}; 0.413_{\pm 0.024} -> 0.455_{\pm 0.030}; 0.558_{\pm 0.027} -> 0.601_{\pm 0.097} | Improved |
| Increase hierarchy depth from 3 to 4 (ablation) | HCRL1 and HCRL2 · MNIST, CIFAR-100, RCV1_v2, and 20Newsgroups | NLL: HCRL1: 209.59 -> 212.31; 1864.69 -> 1860.22; 2562.79 -> 2555.84; 2732.10 -> 2727.49. HCRL2: 203.24 -> 203.91; 1843.40 -> | Mixed |
Closest papers by meaning
- Objective-Based Hierarchical Clustering of Deep Embedding Vectors (AAAI 2021 · similarity 0.90)
- Hierarchical Correlation Clustering and Tree Preserving Embedding (CVPR 2024 · similarity 0.88)
- From Logits to Hierarchies: Hierarchical Clustering made Simple (ICML 2025 · similarity 0.88)
- Classification Fields: Arbitrarily Fine Recursive Hierarchical Clustering From Few Examples (NeurIPS 2026 · similarity 0.86)
- Clustering Properties of Self-Supervised Learning (ICML 2025 · similarity 0.85)
- Self-Paced Unified Representation Learning for Hierarchical Multi-Label Classification (AAAI 2024 · similarity 0.85)
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.
