Stein Variational Gradient Descent: A General Purpose Bayesian Inference Algorithm
This provides a scalable and efficient method for Bayesian inference, applicable across various domains, though it builds on existing kernelized Stein discrepancy theory.
The paper tackles the problem of Bayesian inference by proposing Stein Variational Gradient Descent, a general-purpose algorithm that iteratively transports particles to match target distributions using functional gradient descent, and it shows competitive performance with state-of-the-art methods on real-world models and datasets.
We propose a general purpose variational inference algorithm that forms a natural counterpart of gradient descent for optimization. Our method iteratively transports a set of particles to match the target distribution, by applying a form of functional gradient descent that minimizes the KL divergence. Empirical studies are performed on various real world models and datasets, on which our method is competitive with existing state-of-the-art methods. The derivation of our method is based on a new theoretical result that connects the derivative of KL divergence under smooth transforms with Stein's identity and a recently proposed kernelized Stein discrepancy, which is of independent interest.