Research map / Probabilistic and scientific modeling / Variational and generative sampling
Understanding and Accelerating Particle-Based Variational Inference
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 |
|---|---|---|---|
| apply WNes acceleration to SVGD | SVGD · Kin8nm | Avg. Test RMSE (×10^-2): 8.4$\pm$0.2 -> 6.9$\pm$0.1 | Helped |
| apply WNes acceleration to Blob | Blob · Kin8nm | Avg. Test RMSE (×10^-2): 8.2$\pm$0.2 -> 7.0$\pm$0.2 | Helped |
| apply WNes acceleration to GFSD | GFSD · Kin8nm | Avg. Test RMSE (×10^-2): 8.0$\pm$0.3 -> 6.9$\pm$0.1 | Helped |
| apply WNes acceleration to GFSF | GFSF · Kin8nm | Avg. Test RMSE (×10^-2): 8.3$\pm$0.2 -> 6.8$\pm$0.1 | Helped |
| apply WNes acceleration to SVGD | SVGD · Kin8nm | Avg. Test LL: 1.042$\pm$0.016 -> 1.171$\pm$0.014 | Helped |
| apply WNes acceleration to Blob | Blob · Kin8nm | Avg. Test LL: 1.079$\pm$0.021 -> 1.168$\pm$0.014 | Helped |
| apply WNes acceleration to GFSD | GFSD · Kin8nm | Avg. Test LL: 1.087$\pm$0.029 -> 1.173$\pm$0.016 | Helped |
| apply WNes acceleration to GFSF | GFSF · Kin8nm | Avg. Test LL: 1.044$\pm$0.016 -> 1.193$\pm$0.014 | Helped |
| use WNes rather than WAG acceleration | ParVIs · Bayesian logistic regression, Covertype | test accuracy | Helped |
| use WNes rather than WAG acceleration | ParVIs · Kin8nm | Avg. Test RMSE (×10^-2) | Mixed |
| use WNes rather than WAG acceleration | ParVIs · Kin8nm | Avg. Test LL | Mixed |
| use the HE bandwidth method instead of the median method | kernel-based ParVIs · Toy experiments | particle representativeness | Helped |
Closest papers by meaning
- Variance Reduction and Quasi-Newton for Particle-Based Variational Inference (ICML 2020 · similarity 0.95)
- Particle Semi-Implicit Variational Inference (NeurIPS 2024 · similarity 0.92)
- Functional Gradient Flows for Constrained Sampling (NeurIPS 2024 · similarity 0.90)
- GAD-PVI: A General Accelerated Dynamic-Weight Particle-Based Variational Inference Framework (AAAI 2024 · similarity 0.90)
- Theoretical Guarantees for Variational Inference with Fixed-Variance Mixture of Gaussians (ICML 2024 · similarity 0.89)
- To discretize continually: Mean shift interacting particle systems for Bayesian inference (NeurIPS 2026 · similarity 0.88)
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.