Poster
Stein Variational Gradient Descent With Matrix-Valued Kernels
Dilin Wang · Ziyang Tang · Chandrajit Bajaj · Qiang Liu

Wed Dec 11th 05:00 -- 07:00 PM @ East Exhibition Hall B + C #192

Stein variational gradient descent (SVGD) is a particle-based inference algorithm that leverages gradient information for efficient approximate inference. In this work, we enhance SVGD by leveraging preconditioning matrices, such as the Hessian and Fisher information matrix, to incorporate geometric information into SVGD updates. We achieve this by presenting a generalization of SVGD that replaces the scalar-valued kernels in vanilla SVGD with more general matrix-valued kernels. This yields a significant extension of SVGD, and more importantly, allows us to flexibly incorporate various preconditioning matricesto accelerate the exploration in the probability landscape. Empirical results show that our method outperforms vanilla SVGD and a variety of baseline approaches over a range of real-world Bayesian inference tasks.

Author Information

Dilin Wang (UT Austin)
Ziyang Tang (UT Austin)
Chandrajit Bajaj (The University of Texas at Austin)
Qiang Liu (UT Austin)

More from the Same Authors