Research map / Probabilistic and scientific modeling / Variational and generative sampling
A Heavy-Tailed Algebra for Probabilistic Programming
Feynman T. Liang, Liam Hodgkinson, Michael W. Mahoney
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 |
|---|---|---|---|
| replace Normal base distribution with GGA base distribution in flow | Normal Flow · VI, Cauchy target | Pareto k-hat: 0.22 -> 0.091 | Improved |
| replace Normal base with GGA base in flow | Normal Flow · VI, Chi2 target | Pareto k-hat: 0.61 -> 0.089 | Improved |
| replace Normal base with GGA base in flow | Normal Flow · VI, Normal target | Pareto k-hat: 0.21 -> 0.2 | No clear effect |
| replace Normal base with GGA base in flow | Normal Flow · VI, IG target | Pareto k-hat: 8.2 -> 2.9 | Mixed |
| replace Normal base with GGA base in flow | Normal Flow · VI, StudentT target | Pareto k-hat: 1.0 -> 1.3 | Hurt |
| replace Normal affine with GGA affine transform (ablation) | Normal Affine · VI, IG target | Pareto k-hat: 8.6 -> 2.0 | Improved |
| replace Normal affine with GGA affine transform (ablation) | Normal Affine · VI, Chi2 target | Pareto k-hat: 0.57 -> 0.0093 | Improved |
| replace affine map with flow under GGA base (ablation) | GGA Affine · VI, Cauchy target | Pareto k-hat: 0.68 -> 0.091 | Improved |
| replace Normal base with GGA base in flow | Normal Flow · Density estimation, Cauchy target | estimated alpha-hat: 7.7 -> 2.1 | Improved |
| replace Normal base with GGA base in flow | Normal Flow · Density estimation, StudentT target | estimated alpha-hat: 7.7 -> 3.3 | Improved |
| use GGA base instead of fixed Cauchy base | Cauchy Flow · Density estimation, StudentT target | estimated alpha-hat: 2.0 -> 3.3 | Improved |
| use GGA base instead of Cauchy base | Cauchy Flow · Density estimation, Normal target | estimated alpha-hat: 2.9 -> 8.2 | Improved |
Closest papers by meaning
- Models of Heavy-Tailed Mechanistic Universality (ICML 2025 · similarity 0.86)
- Emergence of heavy tails in homogenized stochastic gradient descent (NeurIPS 2024 · similarity 0.85)
- Flexible Tails for Normalizing Flows (ICML 2025 · similarity 0.85)
- A Separation in Heavy-Tailed Sampling: Gaussian vs. Stable Oracles for Proximal Samplers (NeurIPS 2024 · similarity 0.84)
- Generalization Bounds for Heavy-Tailed SDEs through the Fractional Fokker-Planck Equation (ICML 2024 · similarity 0.84)
- Long-Tailed Learning as Multi-Objective Optimization (AAAI 2024 · 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.
