Research map / Probabilistic and scientific modeling / Uncertainty and Gaussian processes
Quantile Graphical Models: a Bayesian Approach
Nilabja Guha, Veera Baladandayuthapani, Bani K. Mallick
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 quantile variational Bayes (tau=.5) instead of GGM (MB) | graph estimation · Example 1(a) (P=30, n=400, nonlinear/heavy-tailed) | FDR: 5.21 -> 0.28 | Improved |
| use quantile variational Bayes (tau=.5) instead of GGM (GLASSO) | graph estimation · Example 1(a) (P=30, n=400, nonlinear/heavy-tailed) | FDR: 14.78 -> 0.28 | Improved |
| add quantiles .3 and .7 to tau={.5} (ablation) | QVB · Example 1(a) (P=30, n=400, nonlinear/heavy-tailed) | e1 / FDR: e1 0.32 -> 0.17; FDR 0.28 -> 0.58 | Mixed |
| use higher quantile tau=.9 instead of tau=.3 (ablation) | QVB-MCMC · Example 3a, n=200 | e_n: 3.88 -> 1.65 | Improved |
| increase sample size from n=200 to n=500 (ablation) | QVB-MCMC tau=.9 · Example 3a | e_n: 1.65 -> 0.29 | Improved |
| use quantiles .2,.5,.8 instead of tau=.5 alone (ablation) | QVB · Example 3a.ii (hierarchical scale) | detecting connections to scale variable X16 | Improved |
| densify quantile grid from step 0.1 to 0.025 (ablation) | QVB · Example 3b (granularity of quantile grid) | FDR: 0.68 -> 0.77 | No clear effect |
| add quantiles .3 and .7 to tau={.5} (ablation) | QVB · Example 3b (granularity of quantile grid) | e1: 0.30 -> 0.11 | Improved |
| use variational Bayes instead of MCMC | QVB · Example 1 setup, n=400 | runtime speedup: 170 and 134 times faster | Improved |
| vary prior variance v from 1 to 100 (ablation) | QVB · Example 1(a) | graph estimate | No clear effect |
Closest papers by meaning
- Bayesian Covariate-Dependent Gaussian Graphical Models with Varying Structure (JMLR 2022 · similarity 0.90)
- The Multiple Quantile Graphical Model (NeurIPS 2016 · similarity 0.89)
- Variational Inference for Sparse and Undirected Models (ICML 2017 · similarity 0.86)
- Joint Structural Estimation of Multiple Graphical Models (JMLR 2016 · similarity 0.86)
- A Generic Family of Graphical Models: Diversity, Efficiency, and Heterogeneity (ICML 2025 · similarity 0.86)
- Learning Local Neighborhoods of Non-Gaussian Graphical Models (AAAI 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.
