Research map / Probabilistic and scientific modeling / Variational and generative sampling
Density Estimation via Discrepancy Based Adaptive Sequential Partition
Dangna Li, Kun Yang, Wing Hung Wong
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 |
|---|---|---|---|
| DSP vs best other density estimator on accuracy | DSP · Gaussian mixture d=2..6 | Hellinger distance | No clear effect |
| DSP vs BSP, KDE, OPT on speed | DSP · Gaussian mixture n=10^5 | CPU time: 230.179 -> 0.242 | Improved |
| DSP vs fastest baseline BSP | DSP · d=6, n=10^4 | CPU time: 21.696 -> 0.127 | Improved |
| DSP vs BSP at high dimension | DSP · d=5, n=10^5 | CPU time: 115.869 -> 0.706 | Improved |
| DSP vs OPT at n=10^4, d=2 | DSP · Gaussian mixture | Hellinger distance: 0.0957 -> 0.0803 | Improved |
| Vary m from 3 to 20 (ablation) | DSP · Gaussian mixture | Hellinger distance | No clear effect |
| Increase sample size n | DSP · simulated densities | approximation error | Improved |
| Use DSP-kmeans instead of k-means++ init | k-means · Road network k=40 | improvement over random init: 11.7 -> 13.62 | Improved |
| Use DSP-kmeans instead of k-means++ init | k-means · Stem cell k=60 | improvement over random init: 6.12 -> 13.7 | Improved |
| Use DSP-kmeans instead of k-means++ init | k-means · US census k=4 | improvement over random init: 47.44 -> 40.44 | Hurt |
| Use DSP-kmeans instead of k-means++ init | k-means · Mouse bone marrow k=4 | improvement over random init: 1.51 -> 0.4 | Hurt |
| Use DSP-kmeans vs PCA-REC init | k-means · Stem cell k=60 | improvement over random init: -7.29 -> 13.7 | Improved |
Closest papers by meaning
- Convergence Rates of a Class of Multivariate Density Estimation Methods Based on Adaptive Partitioning (JMLR 2023 · similarity 0.90)
- Convergence rates of a partition based Bayesian multivariate density estimation method (NeurIPS 2017 · similarity 0.88)
- The Polynomial Stein Discrepancy for Assessing Moment Convergence (ICML 2025 · similarity 0.85)
- Breaking the curse of dimensionality in structured density estimation (NeurIPS 2024 · similarity 0.84)
- To discretize continually: Mean shift interacting particle systems for Bayesian inference (NeurIPS 2026 · similarity 0.83)
- SD-KDE: Score-Debiased Kernel Density Estimation (NeurIPS 2025 · similarity 0.83)
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.
