{"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/riemannian-stein-variational-gradient-descent","title":"Riemannian Stein Variational Gradient Descent for Bayesian Inference","arxiv_id":"1711.11216","date":"2017-11-30","proceeding":null,"authors":["Chang Liu","Jun Zhu"],"abstract":"We develop Riemannian Stein Variational Gradient Descent (RSVGD), a Bayesian\ninference method that generalizes Stein Variational Gradient Descent (SVGD) to\nRiemann manifold. The benefits are two-folds: (i) for inference tasks in\nEuclidean spaces, RSVGD has the advantage over SVGD of utilizing information\ngeometry, and (ii) for inference tasks on Riemann manifolds, RSVGD brings the\nunique advantages of SVGD to the Riemannian world. To appropriately transfer to\nRiemann manifolds, we conceive novel and non-trivial techniques for RSVGD,\nwhich are required by the intrinsically different characteristics of general\nRiemann manifolds from Euclidean spaces. We also discover Riemannian Stein's\nIdentity and Riemannian Kernelized Stein Discrepancy. Experimental results show\nthe advantages over SVGD of exploring distribution geometry and the advantages\nof particle-efficiency, iteration-effectiveness and approximation flexibility\nover other inference methods on Riemann manifolds.","url_abs":"http://arxiv.org/abs/1711.11216v1","url_pdf":"http://arxiv.org/pdf/1711.11216v1.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":"riemannian-stein-variational-gradient-descent","repo_url":"https://github.com/changliu00/Riem-SVGD","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"tf","reach":{"status":"unanswered"}},{"paper_slug":"riemannian-stein-variational-gradient-descent","repo_url":"https://github.com/MindSpore-scientific-2/code-8/tree/main/SVGD","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"mindspore","reach":null}],"tasks":[{"task_slug":"bayesian-inference","task_name":"Bayesian Inference"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1711.11216","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}