Stein Variational Adaptive Importance Sampling

Abstract

We propose a novel adaptive importance sampling algorithm which incorporates Stein variational gradient decent algorithm (SVGD) with importance sampling (IS). Our algorithm leverages the nonparametric transforms in SVGD to iteratively decrease the KL divergence between our importance proposal and the target distribution. The advantages of this algorithm are twofold: first, our algorithm turns SVGD into a standard IS algorithm, allowing us to use standard diagnostic and analytic tools of IS to evaluate and interpret the results; second, we do not restrict the choice of our importance proposal to predefined distribution families like traditional (adaptive) IS methods. Empirical experiments demonstrate that our algorithm performs well on evaluating partition functions of restricted Boltzmann machines and testing likelihood of variational auto-encoders.

References in corpus (1)

• Measuring Sample Quality with Stein's Method

Cited by in corpus (8)

• A Stein variational Newton method
• Stein Variational Gradient Descent Without Gradient
• Advances in Variational Inference
• Inference via low-dimensional couplings
• Semantics Preserving Adversarial Learning
• Stein Self-Repulsive Dynamics: Benefits From Past Samples
• Scalable Approximate Inference and Some Applications
• Two Stochastic Control Problems In Capital Structure and Portfolio Choice