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

Download:

Current browse context:

cs.LG

Change to browse by:

References & Citations

DBLP - CS Bibliography

Bookmark

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

Computer Science > Machine Learning

Title: Wasserstein Distributionally Robust Optimization and Variation Regularization

Abstract: Wasserstein distributionally robust optimization (DRO) has recently achieved empirical success for various applications in operations research and machine learning, owing partly to its regularization effect. Although connection between Wasserstein DRO and regularization has been established in several settings, existing results often require restrictive assumptions, such as smoothness or convexity, that are not satisfied for many problems. In this paper, we develop a general theory on the variation regularization effect of the Wasserstein DRO - a new form of regularization that generalizes total-variation regularization, Lipschitz regularization and gradient regularization. Our results cover possibly non-convex and non-smooth losses and losses on non-Euclidean spaces. Examples include multi-item newsvendor, portfolio selection, linear prediction, neural networks, manifold learning, and intensity estimation for Poisson processes, etc. As an application of our theory of variation regularization, we derive new generalization guarantees for adversarial robust learning.
Comments: The paper is previously titled "Wasserstein Distributional Robustness and Regularization in Statistical Learning"
Subjects: Machine Learning (cs.LG); Optimization and Control (math.OC); Machine Learning (stat.ML)
Cite as: arXiv:1712.06050 [cs.LG]
  (or arXiv:1712.06050v3 [cs.LG] for this version)

Submission history

From: Rui Gao [view email]
[v1] Sun, 17 Dec 2017 02:47:14 GMT (1743kb,D)
[v2] Tue, 26 Dec 2017 15:50:30 GMT (2779kb,D)
[v3] Fri, 30 Oct 2020 17:56:21 GMT (1289kb)

Link back to: arXiv, form interface, contact.