Research map / Graph learning / Graph self-supervised learning
Probabilistic Dimensionality Reduction via Structure Learning
Li Wang, Qi Mao
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 small generative noise parameter gamma=10^-3 instead of gamma=0 (ablation) | SPL · DistortedSShape | embedding structure quality | Improved |
| use large C (less error tolerance) in SPL (ablation) | SPL · DistortedSShape | embedding structure quality | Improved |
| vary gamma and C over wide ranges (ablation) | SPL · DistortedSShape | embedding quality | No clear effect |
| add generative noise model (gamma=10^-3 vs 0) (ablation) | SPL · Teapot images | embedding shape | No clear effect |
| add generative noise model (gamma=10^-3 vs 0) (ablation) | SPL · USPS digit 1 images | embedding smoothness | Improved |
| use SPL instead of MVU | embedding for 1-NN · Iris | accuracy: 0.6533 -> 0.9600 | Improved |
| use SPL instead of KPCA | embedding for 1-NN · CMU-PIE | accuracy: 0.2701 -> 0.9805 | Improved |
| use SPL instead of MEU | embedding for 1-NN · Letter | accuracy: 0.1244 -> 0.9054 | Improved |
| use SPL instead of tSNE | embedding for 1-NN · Satimage | accuracy: 0.9037 -> 0.8938 | Hurt |
| use SPL instead of tSNE | embedding for 1-NN · Letter | accuracy: 0.9134 -> 0.9054 | Hurt |
| use SPL instead of tSNE | embedding for 1-NN · Vehicle | accuracy: 0.6738 -> 0.6915 | Improved |
| use SPL instead of tSNE | embedding for 1-NN · Pendigits | accuracy: 0.9909 -> 0.9943 | Improved |
Closest papers by meaning
- On Probabilistic Embeddings in Optimal Dimension Reduction (JMLR 2025 · similarity 0.89)
- Hierarchical Nearest Neighbor Graph Embedding for Efficient Dimensionality Reduction (CVPR 2022 · similarity 0.88)
- How Low Can You Go? Searching for the Intrinsic Dimensionality of Complex Networks using Metric Node Embeddings (ICLR 2025 · similarity 0.87)
- Hierarchical Correlation Clustering and Tree Preserving Embedding (CVPR 2024 · similarity 0.86)
- Dimension Reduction with Locally Adjusted Graphs (AAAI 2025 · similarity 0.86)
- Partition First, Embed Later: Laplacian-Based Feature Partitioning for Refined Embedding and Visualization of High-Dimensional Data (ICML 2025 · 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.
