BriefGPT.xyz
Oct, 2024
斯坦因变分演化策略
Stein Variational Evolution Strategies
HTML
PDF
Cornelius V. Braun, Robert T. Lange, Marc Toussaint
TL;DR
本研究针对现有斯坦因变分梯度下降方法在无梯度信息情况下的局限性,提出了一种将演化策略与斯坦因变分步骤相结合的新方法。这种方法能够在没有梯度信息的情况下,从未规范化的目标密度中生成高质量样本,并在多个具有挑战性的基准问题中表现出显著的性能提升。
Abstract
Stein Variational Gradient Descent
(SVGD) is a highly efficient method to sample from an
Unnormalized Probability Distribution
. However, the SVGD update relies on gradients of the log-density, which may not alway
→