Research map / Probabilistic and scientific modeling / Variational and generative sampling
Particle-Gibbs Sampling for Bayesian Feature Allocation Models
Alexandre Bouchard-Côté, Andrew Roth
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 |
|---|---|---|---|
| use row-wise Gibbs instead of element-wise Gibbs | Gibbs sampler · Toy LG example (N=100) | log joint probability | Improved |
| use row-wise Gibbs instead of element-wise Gibbs | Gibbs sampler · Linear Gaussian synthetic (FBB K=5) | relative log density | Improved |
| use DPF instead of Gibbs | feature allocation sampler · LG FBB K=20 | relative log density | Improved |
| use DPF instead of Gibbs | feature allocation sampler · LG IBP K=20 | relative log density | Improved |
| use PG instead of Gibbs | feature allocation sampler · LG FBB K=5 | relative log density | No clear effect |
| use DPF instead of Gibbs | feature allocation sampler · LFRM (K=5) | relative log density | Improved |
| use DPF instead of Gibbs | feature allocation sampler · LFRM (K=20 and IBP) | relative log density | No clear effect |
| use PG instead of Gibbs | feature allocation sampler · PyClone model | relative log density | Hurt |
| use DPF instead of Gibbs | feature allocation sampler · PyClone model | B-Cubed F-measure | Improved |
| increase particles to 50 or 100 (ablation) | PG and DPF samplers · LG tuning at 10 seconds | RMSE | Hurt |
| increase particles to 50 or 100 (ablation) | PG and DPF samplers · LG tuning at 1000 seconds | RMSE | No clear effect |
| always resample (threshold 1.0) (ablation) | PG sampler · LG tuning | RMSE | Hurt |
Closest papers by meaning
- Parameter elimination in particle Gibbs sampling (NeurIPS 2019 · similarity 0.89)
- Particle Gibbs Split-Merge Sampling for Bayesian Inference in Mixture Models (JMLR 2017 · similarity 0.89)
- Minibatch Gibbs Sampling on Large Graphical Models (ICML 2018 · similarity 0.88)
- Poisson-Minibatching for Gibbs Sampling with Convergence Rate Guarantees (NeurIPS 2019 · similarity 0.88)
- Stochastic Bouncy Particle Sampler (ICML 2017 · similarity 0.86)
- Improving Gibbs Sampler Scan Quality with DoGS (ICML 2017 · similarity 0.86)
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.
