Approximate Bayesian Inference on Inverse-gamma prior and normal-variance likelihood Wide Meta-prior
0.0003KL DivergenceDT-5
Evaluation results
| Method | Links | |
|---|---|---|
| DT-5Inference Time per 1000 problems (s)=0.016, GMM components=52025.02 | 0.0003 | |
| ACE-5Inference Time per 1000 problems (s)=0.0372025.02 | 0.0048 | |
| DT-2Inference Time per 1000 problems (s)=0.014, GMM components=22025.02 | 0.0058 | |
| SVIInference Time per 1000 problems (s)=1482025.02 | 0.0558 | |
| TABPFNV2Inference Time per 1000 problems (s)=1.522025.02 | 0.1513 | |
| PFN-5000Inference Time per 1000 problems (s)=0.003, n-bucket Riemann distribution=50002025.02 | 0.2935 | |
| PFN-15Inference Time per 1000 problems (s)=0.003, n-bucket Riemann distribution=152025.02 | 331.5 |