{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/kernel-adaptive-metropolis-hastings","title":"Kernel Adaptive Metropolis-Hastings","arxiv_id":"1307.5302","date":"2013-07-19","proceeding":null,"authors":["Dino Sejdinovic","Heiko Strathmann","Maria Lomeli Garcia","Christophe Andrieu","Arthur Gretton"],"abstract":"A Kernel Adaptive Metropolis-Hastings algorithm is introduced, for the\npurpose of sampling from a target distribution with strongly nonlinear support.\nThe algorithm embeds the trajectory of the Markov chain into a reproducing\nkernel Hilbert space (RKHS), such that the feature space covariance of the\nsamples informs the choice of proposal. The procedure is computationally\nefficient and straightforward to implement, since the RKHS moves can be\nintegrated out analytically: our proposal distribution in the original space is\na normal distribution whose mean and covariance depend on where the current\nsample lies in the support of the target distribution, and adapts to its local\ncovariance structure. Furthermore, the procedure requires neither gradients nor\nany other higher order information about the target, making it particularly\nattractive for contexts such as Pseudo-Marginal MCMC. Kernel Adaptive\nMetropolis-Hastings outperforms competing fixed and adaptive samplers on\nmultivariate, highly nonlinear target distributions, arising in both real-world\nand synthetic examples. Code may be downloaded at\nhttps://github.com/karlnapf/kameleon-mcmc.","url_abs":"http://arxiv.org/abs/1307.5302v3","url_pdf":"http://arxiv.org/pdf/1307.5302v3.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"kernel-adaptive-metropolis-hastings","repo_url":"https://github.com/karlnapf/kameleon-mcmc","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"none","reach":null}],"tasks":[],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}