Research map / Probabilistic and scientific modeling / Variational and generative sampling
Bayesian Posterior Approximation via Greedy Particle Optimization
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 MMD-FW instead of SVGD | Bayesian neural net regression · Naval | Avg. Test RMSE: 4.9e-4±7.5e-5 -> 4.2e-4±5.3e-5 | Helped |
| Use MMD-FW instead of SVGD | Bayesian neural net regression · Naval | Avg. Test log likelihood: 6.08±0.11 -> 6.00±0.12 | Hurt |
| Use MMD-FW instead of SVGD | Bayesian neural net regression · Protein | Avg. Test RMSE: 4.51±0.057 -> 4.43±0.035 | Helped |
| Use MMD-FW instead of SVGD | Bayesian neural net regression · Protein | Avg. Test log likelihood: -2.93±0.013 -> -2.91±0.0073 | Helped |
| Use MMD-FW instead of SVGD | Bayesian neural net regression · Year | Avg. Test RMSE: 9.54±0.08 -> 9.50±0.09 | Helped |
| Use MMD-FW instead of SVGD | Bayesian neural net regression · Year | Avg. Test log likelihood: -3.65±0.005 -> -3.65±0.011 | No clear effect |
| Initialize the first particle from the prior rather than near the MAP (ablation) | MMD-FW · Posterior approximation | performance | Hurt |
| Use SVGD to initialize particles before applying MMD-FW (ablation) | MMD-FW · Posterior approximation | MMD | Mixed |
Closest papers by meaning
- Approximate Bayesian Inference with Stein Functional Variational Gradient Descent (ICLR 2023 · similarity 0.93)
- Stein Variational Gradient Descent as Moment Matching (NeurIPS 2018 · similarity 0.93)
- Stein Variational Gradient Descent With Matrix-Valued Kernels (NeurIPS 2019 · similarity 0.93)
- Message Passing Stein Variational Gradient Descent (ICML 2018 · similarity 0.93)
- ELBOing Stein: Variational Bayes with Stein Mixture Inference (ICLR 2025 · similarity 0.92)
- Improved Finite-Particle Convergence Rates for Stein Variational Gradient Descent (ICLR 2025 · 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.