mlsanityData to 7 Oct 2026

Using Perturbation to Improve Goodness-of-Fit Tests based on Kernelized Stein Discrepancy

arXiv 2304.14762 ↗First posted 28 Apr 2023stat.ML cs.LG stat.MEcited by 4

Not on the map. Why?

Abstract (arXiv)

Kernelized Stein discrepancy (KSD) is a score-based discrepancy widely used in goodness-of-fit tests. It can be applied even when the target distribution has an unknown normalising factor, such as in Bayesian analysis. We show theoretically and empirically that the KSD test can suffer from low power when the target and the alternative distributions have the same well-separated modes but differ in mixing proportions. We propose to perturb the observed sample via Markov transition kernels, with respect to which the target distribution is invariant. This allows us to then employ the KSD test on the perturbed sample. We provide numerical evidence that with suitably chosen transition kernels the proposed approach can lead to substantially higher power than the KSD test.

Key facts. Cited by 4 papers in the map (data through 7 Oct 2026). Builds on 24 of its 63 references that are in the map. First posted on arXiv 28 Apr 2023.

Builds on

24 of 63 references.

References · 24

  1. 2019Generative Modeling by Estimating Gradients of the Data Distributioncited by 4,067
  2. 2016Stein Variational Gradient Descent: A General Purpose Bayesian Inference Algorithmcited by 498
  3. 2016A Kernelized Stein Discrepancy for Goodness-of-fit Tests and Model Evaluationcited by 236
  4. 2016A Kernel Test of Goodness of Fitcited by 143
  5. 2017Measuring Sample Quality with Kernelscited by 119
  6. 2015Measuring Sample Quality with Stein's Methodcited by 94
  7. 2021MMD Aggregated Two-Sample Testcited by 74
  8. 2021Robust Generalised Bayesian Inference for Intractable Likelihoodscited by 53
  9. 2017A Linear-Time Kernel Goodness-of-Fit Testcited by 41
  10. 2020Blindness of score-based methods to isolated components and mixing proportionscited by 33
  11. 2021Stein's Method Meets Computational Statistics: A Review of Some Recent Developmentscited by 32
  12. 2016Measuring Sample Quality with Diffusionscited by 25
12 more
  1. 2018Spread Divergencecited by 23
  2. 2020Sliced Kernelized Stein Discrepancycited by 23
  3. 2018A Framework for Adaptive MCMC Targeting Multimodal Distributionscited by 21
  4. 2022KSD Aggregated Goodness-of-fit Testcited by 21
  5. 2019The Barker proposal: combining robustness and efficiency in gradient-based MCMCcited by 20
  6. 2022Towards Healing the Blindness of Score Matchingcited by 13
  7. 2022Controlling Moments with Kernel Stein Discrepanciescited by 13
  8. 2020The reproducing Stein kernel approach for post-hoc corrected samplingcited by 10
  9. 2014Gaussian-binary Restricted Boltzmann Machines on Modeling Natural Image Statisticscited by 7
  10. 2020A fresh take on 'Barker dynamics' for MCMCcited by 6
  11. 2021Active Slices for Sliced Stein Discrepancycited by 3
  12. 2016A Repelling-Attracting Metropolis Algorithm for Multimodalitycited by 2

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2023: 1; 2024: 1; 2025: 1; 2026: 1.

4 papers.

  1. 2023Energy Discrepancies: A Score-Independent Loss for Energy-Based Modelscited by 9
  2. 2024On the Robustness of Kernel Goodness-of-Fit Testscited by 7
  3. 2026Kernel Tests of Equivalencecited by 2
  4. 2025Correcting Mode Proportion Bias in Generalized Bayesian Inference via a Weighted Kernel Stein Discrepancycited by 0