Pith. sign in

REVIEW 3 cited by

Robust Wasserstein Profile Inference and Applications to Machine Learning

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 1610.05627 v4 pith:6RCONVX3 submitted 2016-10-18 math.ST stat.TH

classification math.STstat.TH
keywords wassersteininferencelearningmachinerobustdistancesempiricalestimators
verification ladder T0 review T1 audit T2 compute T3 formal

Signed reviews

No signed human review yet.

0 comments
read the original abstract

We show that several machine learning estimators, including square-root LASSO (Least Absolute Shrinkage and Selection) and regularized logistic regression can be represented as solutions to distributionally robust optimization (DRO) problems. The associated uncertainty regions are based on suitably defined Wasserstein distances. Hence, our representations allow us to view regularization as a result of introducing an artificial adversary that perturbs the empirical distribution to account for out-of-sample effects in loss estimation. In addition, we introduce RWPI (Robust Wasserstein Profile Inference), a novel inference methodology which extends the use of methods inspired by Empirical Likelihood to the setting of optimal transport costs (of which Wasserstein distances are a particular case). We use RWPI to show how to optimally select the size of uncertainty regions, and as a consequence, we are able to choose regularization parameters for these machine learning estimators without the use of cross validation. Numerical experiments are also given to validate our theoretical findings.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 3 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Tractable Reformulations of Distributionally Robust Two-stage Stochastic Programs with $\infty-$Wasserstein Distance

    math.OC 2019-08 conditional novelty 6.0 of 10

    Under sign conditions on the technology matrix, the worst-case expected recourse cost in two-stage distributionally robust programs with infinity-Wasserstein ambiguity is exactly a finite linear or conic program with ...

  2. Wasserstein Distributionally Robust Optimization: Theory and Applications in Machine Learning

    stat.ML 2019-08 accept novelty 3.0 of 10

    Wasserstein distributionally robust optimization yields data-driven decisions that are computable as convex programs and have finite-sample out-of-sample guarantees, and this tutorial unifies the theory with machine l...

  3. Distributionally Robust Optimization: A Review

    math.OC 2019-08 unverdicted

    A broad review of distributionally robust optimization that organizes the literature by ambiguity-set type and connects DRO to robust optimization, risk aversion, chance constraints, and regularization.

Pith tools