# Stein Variational Gradient Descent With Matrix-Valued Kernels

Dilin WangZiyang TangChandrajit BajajQiang Liu

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... (read more)

PDF Abstract