Research map / Learning theory and algorithms / Sample complexity
Statistical Estimation from Dependent Data
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 label-dependency modeling with β instead of setting β = 0 (ablation) | MPLE on Cora · Cora sparse-data | accuracy: 60\pm0.4\% -> 65.8\pm 0.09\% | Helped |
| Add label-dependency modeling with β instead of setting β = 0 (ablation) | MPLE on Citeseer · Citeseer sparse-data | accuracy: 60.2\pm 0.3\% -> 60.9\pm 0.7\% | Helped |
| Add label-dependency modeling with β instead of setting β = 0 (ablation) | MPLE on Pubmed · Pubmed sparse-data | accuracy: 73.3\pm 0.2\% -> 73.3\pm 0.2\% | No clear effect |
| Add label-dependency modeling with β instead of setting β = 0 (ablation) | MPLE on Cora · Cora full-supervised | accuracy: 74.5\pm 1.8 -> \textbf{85.3}\pm 1.7 | Helped |
| Use MPLE-β rather than GCNII-In | Cora · Cora full-supervised | accuracy: \textbf{85.3}\pm 1.3 -> \textbf{85.3}\pm 1.7 | No clear effect |
| Add label-dependency modeling with β instead of setting β = 0 (ablation) | MPLE on Citeseer · Citeseer full-supervised | accuracy: 72.3\pm1.7 -> \textbf{76.3}\pm1.0 | Helped |
| Use MPLE-β rather than GCNII-In | Citeseer · Citeseer full-supervised | accuracy: 68.6 \pm 0.3 -> \textbf{76.3}\pm1.0 | Helped |
| Add label-dependency modeling with β instead of setting β = 0 (ablation) | MPLE on Pubmed · Pubmed full-supervised | accuracy: 87.3\pm 0.2 -> \textbf{89.0}\pm 0.2 | Helped |
| Use MPLE-β rather than GCNII-In | Pubmed · Pubmed full-supervised | accuracy: 83.3 \pm 0.6 -> \textbf{89.0}\pm 0.2 | Helped |
| Increase training data while modeling label dependencies | MPLE-β versus MPLE-0 · Cora, Citeseer, and Pubmed increasing-training-data experiments | accuracy | Helped |
Closest papers by meaning
- A Flexible Framework for Nonparametric Graphical Modeling that Accommodates Machine Learning (ICML 2020 · similarity 0.86)
- InfoAtlas: A Foundation Model for Zero-Shot Statistical Dependence Estimate (ICML 2026 · similarity 0.86)
- Graph Structure Inference with BAM: Neural Dependency Processing via Bilinear Attention (NeurIPS 2024 · similarity 0.84)
- CoInD: Enabling Logical Compositions in Diffusion Models (ICLR 2025 · similarity 0.84)
- Learning Local Neighborhoods of Non-Gaussian Graphical Models (AAAI 2025 · similarity 0.84)
- Fast Wasserstein rates for estimating probability distributions of probabilistic graphical models (NeurIPS 2026 · similarity 0.84)
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.