Research map / Optimization / Optimizers
Projected Stein Variational Gradient Descent
Peng Chen, Omar Ghattas
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 |
|---|---|---|---|
| project parameters to low-dimensional gradient-informed subspace (pSVGD instead of SVGD) | SVGD · Bayesian logistic regression, Arcene (d=10,000) | training time (s): 477 seconds -> 201 seconds | Improved |
| project parameters to low-dimensional subspace (pSVGD instead of SVGD) | SVGD · Bayesian logistic regression, Arcene | prediction accuracy | Improved |
| project parameters to low-dimensional subspace (pSVGD instead of SVGD) | SVGD · elliptic PDE inverse problem, d from 289 to 16,641 | RMSE of pointwise sample variance | Improved |
| project parameters to low-dimensional subspace (pSVGD instead of SVGD) | SVGD · conditional diffusion process, d=100 | posterior mean closeness to true parameter | Improved |
| increase parameter dimension, number of samples, data points | pSVGD · elliptic PDE inverse problem | eigenvalue decay / convergence | No clear effect |
| add more processor cores K for same work (KN=1024 samples) | pSVGD · elliptic PDE inverse problem | CPU time | Improved |
Closest papers by meaning
- Stein Variational Gradient Descent With Matrix-Valued Kernels (NeurIPS 2019 · similarity 0.92)
- Projected Stein Variational Newton: A Fast and Scalable Bayesian Inference Method in High Dimensions (NeurIPS 2019 · similarity 0.92)
- Stein Variational Gradient Descent: A General Purpose Bayesian Inference Algorithm (NeurIPS 2016 · similarity 0.91)
- Approximate Bayesian Inference with Stein Functional Variational Gradient Descent (ICLR 2023 · similarity 0.91)
- ELBOing Stein: Variational Bayes with Stein Mixture Inference (ICLR 2025 · similarity 0.90)
- Bayesian Online Natural Gradient (BONG) (NeurIPS 2024 · similarity 0.87)
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.
