Research map / Optimization / Optimizers
On the geometry of Stein variational gradient descent
Andrew Duncan, Nikolas Nüsken, Lukasz Szpruch
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 |
|---|---|---|---|
| decrease kernel smoothness parameter p from 2 to 1 in the exponential-power kernel (ablation) | SVGD with k_{p,sigma}, 500 particles · 1D four-component Gaussian mixture | particle approximation accuracy (histogram, Wasserstein-1) | Improved |
| decrease p further below 1 (to 1/2 and smaller) (ablation) | SVGD with k_{p,sigma}, 500 particles · 1D four-component Gaussian mixture | Wasserstein-1 error and gradient evaluations | Hurt |
| use p=1 instead of p=2 or p=1/2 at fixed p, measured per gradient evaluation (ablation) | SVGD with k_{p,sigma} · 1D four-component Gaussian mixture | Wasserstein-1 error vs gradient evaluations | Improved |
| anneal p from p0=2 to p1=1/2 over time (log-linear schedule) instead of fixed p (ablation) | SVGD with k_{p,sigma} · 1D four-component Gaussian mixture | Wasserstein-1 error | Mixed |
| use p=0.5 instead of p=2 or p=1 (ablation) | SVGD with k_{p,sigma}, 500 particles · 2D six-component Gaussian mixture | Wasserstein-1 error | Improved |
| use non-translation-invariant weighted Matern kernel with adapted tails | SVGD · numerical studies | computational cost vs accuracy | No clear effect |
Closest papers by meaning
- Stein Variational Gradient Descent: A General Purpose Bayesian Inference Algorithm (NeurIPS 2016 · similarity 0.94)
- Stein Variational Gradient Descent With Matrix-Valued Kernels (NeurIPS 2019 · similarity 0.93)
- Bayesian Posterior Approximation via Greedy Particle Optimization (AAAI 2019 · similarity 0.93)
- Improved Finite-Particle Convergence Rates for Stein Variational Gradient Descent (ICLR 2025 · similarity 0.91)
- ELBOing Stein: Variational Bayes with Stein Mixture Inference (ICLR 2025 · similarity 0.90)
- Functional Gradient Flows for Constrained Sampling (NeurIPS 2024 · similarity 0.89)
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.
