Coverage tests for simulation-based inference of f_NL can pass while the posteriors are underconfident in the tails and sometimes yield weaker constraints than using power spectrum or bispectrum alone.
Proceedings of the 40th International Conference on Machine Learning , pages =
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
astro-ph.CO 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Coverage is not enough: Frequentist tests of simulation-based inference for primordial non-Gaussianity
Coverage tests for simulation-based inference of f_NL can pass while the posteriors are underconfident in the tails and sometimes yield weaker constraints than using power spectrum or bispectrum alone.