Research map / Probabilistic and scientific modeling / Variational and generative sampling
Geodesic Slice Sampling on the Sphere
Michael Habeck, Mareike Hasenpflug, Shantanu Kodgirwar, Daniel Rudolf
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 geoSSS (reject) instead of geoSSS (shrink) | geodesic slice sampling · Rigid registration AK | iterations to 100% success: 1500 -> 200 | Improved |
| use geoSSS instead of RWMH or HMC | MCMC sampler · Rigid registration AK | success rate: 3-7% -> 100% | Improved |
| use geoSSS (shrink) over geoSSS (reject) when measured by computation time | geodesic slice sampling · Rigid registration AK | success vs computation time | Improved |
| use geoSSS instead of RWMH and HMC on multimodal target | MCMC sampler · vMF mixture d=10 K=5 kappa=100 | mode visit KL divergence | Improved |
| use geoSSS (reject) instead of shrink | geodesic slice sampling · vMF mixture kappa 50 to 500 | relative ESS | Improved |
| use geoSSS (reject) instead of shrink | geodesic slice sampling · vMF mixture kappa 50 to 500 | rejections per step: 17 -> 60 vs 4 -> 6 | Hurt |
| increase concentration kappa from 50 to 500 (ablation) | all samplers · vMF mixture d=10 K=5 | relative ESS | Hurt |
| use geoSSS (shrink) instead of mixture-MH tuned alpha | sampler · vMF mixture | KL divergence and ESS | No clear effect |
| use HMC instead of geoSSS | sampler · Curved vMF d=5 kappa 500 | relative ESS | Improved |
| increase dimension from 3 to 24 (ablation) | geoSSS and RWMH · Curved vMF kappa=500 | relative ESS | Hurt |
| use geoSSS or HMC instead of RWMH | sampler · Curved vMF 2-sphere kappa=500 | ACF, KL, jump distance | Improved |
Closest papers by meaning
- Stochastic Gradient Geodesic MCMC Methods (NeurIPS 2016 · similarity 0.90)
- Sampling from Binary Quadratic Distributions via Stochastic Localization (ICML 2025 · similarity 0.85)
- Stochastic Localization via Iterative Posterior Sampling (ICML 2024 · similarity 0.85)
- Faster Sampling from Log-Concave Densities over Polytopes via Efficient Linear Solvers (ICLR 2024 · similarity 0.84)
- Linear Spherical Sliced Optimal Transport: A Fast Metric for Comparing Spherical Data (ICLR 2025 · similarity 0.84)
- Spherical Flows for Sampling Categorical Data (NeurIPS 2026 · 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.
