W-Kernel and Its Principal Space for Frequentist Evaluation of Bayesian Estimators
stat.ME, cond-mat.stat-mech, stat.ML
Submitted: 2023-11-21
Updated: 2026-09-06
Comments: Minor revisions to figures and text, with additions including a Note Added in Proof. Essentially identical in content to the published version, except for editorial changes, notably equation numbering. Supplementary Material appended; references merged
Journal ref: Neural Computation (2026) 38(8) 1408-1468
DOI: 10.1162/NECO.a.1547
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Terminology
Sources
- The Bayesian Infinitesimal Jackknife for Variance
- A Higher-Order Swiss Army Infinitesimal Jackknife
- Posterior covariance information criterion for general loss functions
- The Memory Perturbation Equation: Understanding Model's Sensitivity to Data
- Learning Representation from Neural Fisher Kernel with Low-rank Approximation
- Prior Intensified Information Criterion
Related papers
- Doubly robust inference via calibration
- Bayesian Empirical Bayes: Simultaneous Inference from Probabilistic Symmetries
- Flexible Nonparametric Inference for Causal Effects under the Front-Door Model
- Deployment of AI-Assisted Interventions: Capacity Constraints and Noisy Compliance
- A Survey on Archetypal Analysis
- Dynamic Spatial Bayesian Machine Learning Model: Applications to Intergenerational Economic Mobility and Geographic Income Inequality in the United States