Timezone: »

Kernel Stein Discrepancy Descent
Anna Korba · Pierre-Cyril Aubin-Frankowski · Szymon Majewski · Pierre Ablin

Wed Jul 21 07:00 AM -- 07:20 AM (PDT) @ None
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.

Author Information

Anna Korba (CREST/ENSAE)
Pierre-Cyril Aubin-Frankowski (MINES ParisTech)
Szymon Majewski (Ecole Polytechnique)
Pierre Ablin (CNRS and ENS)

Related Events (a corresponding poster, oral, or spotlight)

More from the Same Authors