arXiv stat.ML
· Papers
$lambda$-PSD: Scalable Approximate SNR-Optimised Polynomial Stein Discrepancies
arXiv:2606.26621v1 Announce Type: cross Abstract: Polynomial Stein discrepancies (PSD) provide a scalable alternative to kernel Stein methods for measuring sample quality and goodness-of-fit testing, but their statistical properties remain poorly understood. We show that increasing polynomial degree primarily amplifies