EUVIMEDEUVIMED entdecken
Crossref

Superquantiles at Work: Machine Learning Applications and Efficient Subgradient Computation

Yassine Laguel, Krishna Pillutla, Jérôme Malick, Zaid Harchaoui

Set-Valued and Variational Analysis · 2021-12 · 29 (4) · 967-996

DOI: 10.1007/s11228-021-00609-w ↗

Kurz gefasst

Fachartikel in der Zeitschrift Set-Valued and Variational Analysis, online erschienen am 30.12.2021. Verfasst von Yassine Laguel und 3 weiteren Autor:innen. Die Arbeit wurde bisher 9-mal zitiert (Crossref) und nennt 57 Literaturangaben. Gefördert von: ANR, National Science Foundation.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

R. Tyrell Rockafellar and his collaborators introduced, in a series of works, new regression modeling methods based on the notion of superquantile (or conditional value-at-risk). These methods have been influential in economics, finance, management science, and operations research in general. Recently, they have been subject of a renewed interest in machine learning, to address issues of distributional robustness and fair allocation. In this paper, we review some of these new applications of the superquantile, with references to recent developments. These applications involve nonsmooth superquantile-based objective functions that admit explicit subgradient calculations. To make these superquantile-based functions amenable to the gradient-based algorithms popular in machine learning, we show how to smooth them by infimal convolution and detail numerical procedures to compute the gradients of the smooth approximations. We put the approach into perspective by comparing it to other smoothing techniques and by illustrating it on toy examples.

Förderung

  • ANR
  • National Science Foundation

Zitieren

APA

Laguel, Y., Pillutla, K., Malick, J., & Harchaoui, Z. (2021). Superquantiles at Work: Machine Learning Applications and Efficient Subgradient Computation. Set-Valued and Variational Analysis, 29(4), 967-996. https://doi.org/10.1007/s11228-021-00609-w

Vancouver

Laguel Y, Pillutla K, Malick J, Harchaoui Z. Superquantiles at Work: Machine Learning Applications and Efficient Subgradient Computation. Set-Valued and Variational Analysis. 2021;29(4):967-996. doi:10.1007/s11228-021-00609-w

BibTeX
@article{Laguel2021,
  author = {Laguel, Yassine and Pillutla, Krishna and Malick, Jérôme and Harchaoui, Zaid},
  title = {Superquantiles at Work: Machine Learning Applications and Efficient Subgradient Computation},
  journal = {Set-Valued and Variational Analysis},
  year = {2021},
  volume = {29},
  number = {4},
  pages = {967--996},
  doi = {10.1007/s11228-021-00609-w}
}
Kostenlos starten →