Preprint · 2014 · Open access

Theory of Convex Optimization for Machine Learning.

Sébastien Bubeck

arXiv (Cornell University) · Published 20 May 2014

Read the free full textCite this paper

In brief

“Theory of Convex Optimization for Machine Learning.” is a 2014 preprint by Sébastien Bubeck, published in arXiv (Cornell University). OpenAlex counts 58 citations as of 16 November 2026. It is open access: the full text is free to read at arXiv (Cornell University).

Citations
58
Most cited in
2016 (10)
References
46
Authors
1

Abstract

This monograph presents the main mathematical ideas in convex optimization. Starting from the fundamental theory of black-box optimization, the material progresses towards recent advances in structural optimization and stochastic optimization. Our presentation of black-box optimization, strongly influenced by the seminal book of Nesterov, includes the analysis of the Ellipsoid Method, as well as (accelerated) gradient descent schemes. We also pay special attention to non-Euclidean settings (relevant algorithms include Frank-Wolfe, Mirror Descent, and Dual Averaging) and discuss their relevance in machine learning. We provide a gentle introduction to structural optimization with FISTA (to optimize a sum of a smooth and a simple non-smooth term), Saddle-Point Mirror Prox (Nemirovski's alternative to Nesterov's smoothing), and a concise description of Interior Point Methods. In stochastic optimization we discuss Stochastic Gradient Descent, mini-batches, Random Coordinate Descent, and sublinear algorithms. We also briefly touch upon convex relaxation of combinatorial problems and the use of randomness to round solutions, as well as random walks based methods.

Cite this paper

Formatted from the catalogue record. Check names, capitals and page numbers against the paper before you submit.

APA 7

Bubeck, S. (2014). Theory of Convex Optimization for Machine Learning. arXiv (Cornell University). https://arxiv.org/pdf/1405.4980

MLA 9

Bubeck, Sébastien. “Theory of Convex Optimization for Machine Learning.” arXiv (Cornell University), 2014. https://arxiv.org/pdf/1405.4980.

Chicago

Bubeck, Sébastien. 2014. “Theory of Convex Optimization for Machine Learning.” arXiv (Cornell University). https://arxiv.org/pdf/1405.4980.

Harvard

Bubeck, S. (2014) ‘Theory of Convex Optimization for Machine Learning.’, arXiv (Cornell University). Available at: https://arxiv.org/pdf/1405.4980.

IEEE

S. Bubeck, “Theory of Convex Optimization for Machine Learning.,” arXiv (Cornell University), 2014.

BibTeX

@misc{bubeck2014theory,
  title     = {{Theory of Convex Optimization for Machine Learning.}},
  author    = {Bubeck, Sébastien},
  howpublished = {arXiv (Cornell University)},
  year      = {2014},
  url       = {https://arxiv.org/pdf/1405.4980}
}

Download for a reference manager: BibTeX (.bib) · RIS for Zotero, Mendeley and EndNote (.ris)

Citations by year

YearCitationsShare of the best year
201588
20161010
201766
201833
201966
202055
202166
202222
202355
202444

Topic and keywords

Physical Sciences › Computer Science › Artificial Intelligence › Stochastic Gradient Optimization Techniques

  • convex optimization
  • black-box optimization
  • ellipsoid method
  • accelerated gradient descent
  • Frank-Wolfe algorithm
  • mirror descent
  • dual averaging
  • FISTA

Work with this paper

Most-cited papers in Computer Science

All highly cited Computer Science papers

Questions about this paper

How do I cite this paper in APA style?

Bubeck, S. (2014). Theory of Convex Optimization for Machine Learning. arXiv (Cornell University). https://arxiv.org/pdf/1405.4980

How many times has this paper been cited?

58 times, according to OpenAlex on 16 November 2026. Google Scholar and Scopus count different sets of sources, so their numbers differ.

Is this paper free to read?

Yes. It is open access: read the free full text at arXiv (Cornell University).

Who wrote this paper?

Sébastien Bubeck.

Bibliographic data from OpenAlex (CC0), last checked 16 November 2026. ResearchGuru does not host this paper; the links go to the publisher or the repository that does.