Stein Variational Gradient Descent as Moment Matching
arXiv:1810.11693
Abstract
Stein variational gradient descent (SVGD) is a non-parametric inference algorithm that evolves a set of particles to fit a given distribution of interest. We analyze the non-asymptotic properties of SVGD, showing that there exists a set of functions, which we call the Stein matching set, whose expectations are exactly estimated by any set of particles that satisfies the fixed point equation of SVGD. This set is the image of Stein operator applied on the feature maps of the positive definite kernel used in SVGD. Our results provide a theoretical framework for analyzing the properties of SVGD with different kernels, shedding insight into optimal kernel choice. In particular, we show that SVGD with linear kernels yields exact estimation of means and variances on Gaussian distributions, while random Fourier features enable probabilistic bounds for distributional approximation. Our results offer a refreshing view of the classical inference problem as fitting Stein's identity or solving the Stein equation, which may motivate more efficient algorithms.
arXiv commentConference on Neural Information Processing Systems (NIPS) 2018
Cited by in corpus (9)
- Physics-Informed Probabilistic Learning of Linear Embeddings of Non-linear Dynamics With Guaranteed Stability
- On the geometry of Stein variational gradient descent
- A Non-Asymptotic Analysis for Stein Variational Gradient Descent
- Black-box Off-policy Estimation for Infinite-Horizon Reinforcement Learning
- Posterior inference unchained with EL_2O
- Parametric Fokker-Planck equation
- Stein Self-Repulsive Dynamics: Benefits From Past Samples
- Stochastic Particle-Optimization Sampling and the Non-Asymptotic Convergence Theory
- Stabilizing Training of Generative Adversarial Nets via Langevin Stein Variational Gradient Descent