Pith. sign in

Sliced Kernelized Stein Discrepancy

1 Pith paper cite this work. Polarity classification is still indexing.

1 Pith paper citing it
abstract

Kernelized Stein discrepancy (KSD), though being extensively used in goodness-of-fit tests and model learning, suffers from the curse-of-dimensionality. We address this issue by proposing the sliced Stein discrepancy and its scalable and kernelized variants, which employ kernel-based test functions defined on the optimal one-dimensional projections. When applied to goodness-of-fit tests, extensive experiments show the proposed discrepancy significantly outperforms KSD and various baselines in high dimensions. For model learning, we show its advantages over existing Stein discrepancy baselines by training independent component analysis models with different discrepancies. We further propose a novel particle inference method called sliced Stein variational gradient descent (S-SVGD) which alleviates the mode-collapse issue of SVGD in training variational autoencoders.

fields

cs.LG 1

years

2024 1

verdicts

CONDITIONAL 1

representative citing papers

Path-Guided Particle-based Sampling

cs.LG · 2024-12-04 · conditional · novelty 5.0

PGPS trains a neural velocity field to transport particles along a log-weighted shrinkage density path, giving a Wasserstein error bound of O(delta) + O(sqrt(h)) and improved mode seeking in Bayesian inference.

citing papers explorer

Showing 1 of 1 citing paper.

  • Path-Guided Particle-based Sampling cs.LG · 2024-12-04 · conditional · none · ref 2020 · internal anchor

    PGPS trains a neural velocity field to transport particles along a log-weighted shrinkage density path, giving a Wasserstein error bound of O(delta) + O(sqrt(h)) and improved mode seeking in Bayesian inference.