Kernel Adaptive Metropolis-Hastings

Abstract

A Kernel Adaptive Metropolis-Hastings algorithm is introduced, for the purpose of sampling from a target distribution with strongly nonlinear support. The algorithm embeds the trajectory of the Markov chain into a reproducing kernel Hilbert space (RKHS), such that the feature space covariance of the samples informs the choice of proposal. The procedure is computationally efficient and straightforward to implement, since the RKHS moves can be integrated out analytically: our proposal distribution in the original space is a normal distribution whose mean and covariance depend on where the current sample lies in the support of the target distribution, and adapts to its local covariance structure. Furthermore, the procedure requires neither gradients nor any other higher order information about the target, making it particularly attractive for contexts such as Pseudo-Marginal MCMC. Kernel Adaptive Metropolis-Hastings outperforms competing fixed and adaptive samplers on multivariate, highly nonlinear target distributions, arising in both real-world and synthetic examples. Code may be downloaded at https://github.com/karlnapf/kameleon-mcmc.

Proceedings of the 31st International Conference on Machine Learning, Beijing, China, 2014; JMLR: W&CP volume 32(2)

References in corpus (2)

• The pseudo-marginal approach for efficient Monte Carlo computations
• Slice sampling covariance hyperparameters of latent Gaussian models

Cited by in corpus (14)

• Kernel Mean Embedding of Distributions: A Review and Beyond
• Learning Decentralized Controllers for Robot Swarms with Graph Neural Networks
• Eigendecompositions of Transfer Operators in Reproducing Kernel Hilbert Spaces
• Gradient-free Hamiltonian Monte Carlo with Efficient Kernel Exponential Families
• K2-ABC: Approximate Bayesian Computation with Kernel Embeddings
• Stein Variational Gradient Descent Without Gradient
• A Wild Bootstrap for Degenerate Kernel Tests
• Adaptive Monte Carlo augmented with normalizing flows
• Walsh-Hadamard Variational Inference for Bayesian Deep Learning
• Adaptive Path Sampling in Metastable Posterior Distributions
• Grasp Learning by Sampling from Demonstration
• Active and Transfer Learning of Grasps by Sampling from Demonstration
• Adaptive MCMC via Combining Local Samplers
• Scalable Approximate Inference and Some Applications