We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:


Current browse context:


Change to browse by:

References & Citations


(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Statistics > Machine Learning

Title: Kernel Mean Embedding of Distributions: A Review and Beyond

Abstract: A Hilbert space embedding of a distribution---in short, a kernel mean embedding---has recently emerged as a powerful tool for machine learning and inference. The basic idea behind this framework is to map distributions into a reproducing kernel Hilbert space (RKHS) in which the whole arsenal of kernel methods can be extended to probability measures. It can be viewed as a generalization of the original "feature map" common to support vector machines (SVMs) and other kernel methods. While initially closely associated with the latter, it has meanwhile found application in fields ranging from kernel machines and probabilistic modeling to statistical inference, causal discovery, and deep learning. The goal of this survey is to give a comprehensive review of existing work and recent advances in this research area, and to discuss the most challenging issues and open problems that could lead to new research directions. The survey begins with a brief introduction to the RKHS and positive definite kernels which forms the backbone of this survey, followed by a thorough discussion of the Hilbert space embedding of marginal distributions, theoretical guarantees, and a review of its applications. The embedding of distributions enables us to apply RKHS methods to probability measures which prompts a wide range of applications such as kernel two-sample testing, independent testing, and learning on distributional data. Next, we discuss the Hilbert space embedding for conditional distributions, give theoretical insights, and review some applications. The conditional mean embedding enables us to perform sum, product, and Bayes' rules---which are ubiquitous in graphical model, probabilistic inference, and reinforcement learning---in a non-parametric way. We then discuss relationships between this framework and other related areas. Lastly, we give some suggestions on future research directions.
Comments: 147 pages; this is the final version
Subjects: Machine Learning (stat.ML); Machine Learning (cs.LG)
Journal reference: Foundations and Trends in Machine Learning: Vol. 10: No. 1-2, pp 1-141 (2017)
DOI: 10.1561/2200000060
Cite as: arXiv:1605.09522 [stat.ML]
  (or arXiv:1605.09522v4 [stat.ML] for this version)

Submission history

From: Krikamol Muandet [view email]
[v1] Tue, 31 May 2016 08:23:33 GMT (260kb)
[v2] Thu, 2 Jun 2016 04:25:09 GMT (260kb)
[v3] Wed, 25 Jan 2017 15:05:47 GMT (268kb)
[v4] Sun, 13 Dec 2020 12:45:23 GMT (270kb)

Link back to: arXiv, form interface, contact.