paper-with-me

Papers

Kernel Stein Discrepancy Descent

2021-05-20 · Anna Korba, Pierre-Cyril Aubin-Frankowski, Szymon Majewski, Pierre Ablin

Among dissimilarities between probability distributions, the Kernel Stein Discrepancy (KSD) has received much interest recently. We investigate the properties of its Wasserstein gradient flow to approximate a target probability distribution $\pi$ on $\mathbb{R}^d$, known up to a normalization constant. This leads to a straightforwardly implementable, deterministic score-based method to sample from $\pi$, named KSD Descent, which uses a set of particles to approximate $\pi$. Remarkably, owing to a tractable loss function, KSD Descent can leverage robust parameter-free optimization schemes such as L-BFGS; this contrasts with other popular particle-based schemes such as the Stein Variational Gradient Descent algorithm. We study the convergence properties of KSD Descent and demonstrate its practical relevance. However, we also highlight failure cases by showing that the algorithm can get stuck in spurious local minima.

📄 PDF Abstract BibTeX arXiv:2105.09994

Code (2)

pierreablin/ksddescent 공식 구현 pytorch
louissharrock/coin-svgd pytorch

Similar Papers 제목 키워드 기반

Sliced Kernelized Stein Discrepancy

2020-06-30 · ICLR 2021 1 · Wenbo Gong, Yingzhen Li, José Miguel Hernández-Lobato

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…

Low Stein Discrepancy via Message-Passing Monte Carlo

2025-03-27 · Nathan Kirk, T. Konstantin Rusch, Jakob Zech, Daniela Rus

Message-Passing Monte Carlo (MPMC) was recently introduced as a novel low-discrepancy sampling approach leveraging tools from geometric deep learning. While originally designed for generating uniform point sets, we exten…

Stein Variational Gradient Descent with Multiple Kernel

2021-07-20 · Qingzhong Ai, Shiyu Liu, Lirong He, Zenglin Xu

Stein variational gradient descent (SVGD) and its variants have shown promising successes in approximate inference for complex distributions. In practice, we notice that the kernel used in SVGD-based methods has a decisi…

Computational Efficiency

A Non-Asymptotic Analysis for Stein Variational Gradient Descent

2020-06-17 · NeurIPS 2020 12 · Anna Korba, Adil Salim, Michael Arbel, Giulia Luise 외

We study the Stein Variational Gradient Descent (SVGD) algorithm, which optimises a set of particles to approximate a target probability distribution $\pi\propto e^{-V}$ on $\mathbb{R}^d$. In the population limit, SVGD p…

LEMMA

Minimum Stein Discrepancy Estimators

2019-06-19 · NeurIPS 2019 12 · Alessandro Barp, Francois-Xavier Briol, Andrew B. Duncan, Mark Girolami 외

When maximum likelihood estimation is infeasible, one often turns to score matching, contrastive divergence, or minimum probability flow to obtain tractable parameter estimates. We provide a unifying perspective of these…