Research map / Probabilistic and scientific modeling / Uncertainty and Gaussian processes
Semiparametric Inference Using Fractional Posteriors
Alice L'Huillier, Luke Travis, Ismaël Castillo, Kolyan Ray
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 |
|---|---|---|---|
| apply shift-and-rescale correction to alpha_n-posterior with alpha_n above the threshold (Case 1) | alpha_n-posterior credible set · Case 1 Gaussian white noise | empirical coverage: 1.00 -> 0.95 (length 0.08 -> 0.02) | Improved |
| let alpha_n fall below the theoretical lower bound for shift-and-rescale sets (Case 1) (ablation) | Shift-and-rescale set · Case 1 Gaussian white noise | empirical coverage: 0.95 -> 0.00 | Hurt |
| let alpha_n fall below the theoretical lower bound for shift-and-rescale sets (Case 2) (ablation) | Shift-and-rescale set · Case 2 Gaussian white noise | empirical coverage: 0.95 -> 0.15 | Hurt |
| let alpha_n fall below the theoretical lower bound for shift-and-rescale sets (Case 3) (ablation) | Shift-and-rescale set · Case 3 Gaussian white noise | empirical coverage: 0.94 -> 0.00 | Hurt |
| let alpha_n fall below the theoretical lower bound for shift-and-rescale sets (density estimation) (ablation) | Shift-and-rescale set · Case 4 density estimation | empirical coverage: 0.92 -> 0.31 | Hurt |
| use raw alpha_n-posterior credible set instead of the corrected set (Case 1) | alpha_n-posterior credible set · Case 1 Gaussian white noise | credible set length: 0.02 -> 0.86 | Hurt |
| use raw alpha_n-posterior credible set instead of the corrected set (Case 2) | alpha_n-posterior credible set · Case 2 Gaussian white noise | credible set length: 0.02 -> 0.29 | Hurt |
| use smaller alpha_n (below threshold) in alpha_n-posterior centering (ablation) | alpha_n-posterior centering bias · Case 1 Gaussian white noise | centering bias: -0.00059 -> -0.05331 | Hurt |
Closest papers by meaning
- Asymptotic Consistency of $\alpha$-{R}ényi-Approximate Posteriors (JMLR 2020 · similarity 0.85)
- Variational Inference for SDEs Driven by Fractional Noise (ICLR 2024 · similarity 0.84)
- Variational Learning of Fractional Posteriors (ICML 2025 · similarity 0.84)
- Asymptotic Anytime-Valid Inference for U-statistics (NeurIPS 2026 · similarity 0.83)
- Amortized Structured Stochastic Variational Inference for Gaussian Process Latent Variable Models (NeurIPS 2026 · similarity 0.83)
- Online Bernstein-von Mises theorem (NeurIPS 2026 · similarity 0.83)
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.
