Research map / Optimization / Optimizers
Stein Variational Gradient Descent: A General Purpose Bayesian Inference Algorithm
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 SVGD (20 particles) instead of PBP for Bayesian neural network inference | Bayesian neural networks on UCI regression datasets · UCI regression (9 datasets), test RMSE | Avg. Test RMSE | Helped |
| Use SVGD instead of PBP on Yacht dataset | Bayesian neural networks on UCI Yacht · UCI Yacht, test RMSE | Avg. Test RMSE: 0.000 | No clear effect |
| Use SVGD instead of PBP for Bayesian neural network training time | Bayesian neural networks on UCI regression datasets · UCI regression, average training time | Avg. Time (Secs) | Helped |
| Use a single particle (n=1) instead of many particles | SVGD · Toy 1D Gaussian mixture | qualitative mode recovery | Helped |
| Use parallel SGLD / SVGD with 100 particles vs sequential SGLD on Covertype | Bayesian logistic regression on Covertype · Covertype test accuracy | Testing accuracy | Mixed |
Closest papers by meaning
- Approximate Bayesian Inference with Stein Functional Variational Gradient Descent (ICLR 2023 · 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)
- Bayesian Online Natural Gradient (BONG) (NeurIPS 2024 · similarity 0.90)
- ELBOing Stein: Variational Bayes with Stein Mixture Inference (ICLR 2025 · similarity 0.90)
- Generalized Variational Inference via Optimal Transport (AAAI 2024 · similarity 0.90)
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.