Journal article · 2014 · Open access

Learning to Optimize via Posterior Sampling

Daniel Russo, Benjamin Van Roy

Mathematics of Operations Research, 39(4), 1221–1243 · Published 23 April 2014 · DOI 10.1287/moor.2014.0650

Open PDFView at the publisherCite this paper

In brief

“Learning to Optimize via Posterior Sampling” is a 2014 journal article by Daniel Russo and Benjamin Van Roy, published in Mathematics of Operations Research. OpenAlex counts 825 citations as of 19 November 2026, which puts it in the top 7% of works published in 2014 by citations. It is open access: the full text is free to read at arXiv (Cornell University).

Citations
825
Most cited in
2020 (143)
Field-weighted impact
27.19
References
45
Authors
2

Abstract

This paper considers the use of a simple posterior sampling algorithm to balance between exploration and exploitation when learning to optimize actions such as in multiarmed bandit problems. The algorithm, also known as Thompson Sampling and as probability matching, offers significant advantages over the popular upper confidence bound (UCB) approach, and can be applied to problems with finite or infinite action spaces and complicated relationships among action rewards. We make two theoretical contributions. The first establishes a connection between posterior sampling and UCB algorithms. This result lets us convert regret bounds developed for UCB algorithms into Bayesian regret bounds for posterior sampling. Our second theoretical contribution is a Bayesian regret bound for posterior sampling that applies broadly and can be specialized to many model classes. This bound depends on a new notion we refer to as the eluder dimension, which measures the degree of dependence among action rewards. Compared to UCB algorithm Bayesian regret bounds for specific model classes, our general bound matches the best available for linear models and is stronger than the best available for generalized linear models. Further, our analysis provides insight into performance advantages of posterior sampling, which are highlighted through simulation results that demonstrate performance surpassing recently proposed UCB algorithms.

Cite this paper

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

APA 7

Russo, D., & Van Roy, B. (2014). Learning to Optimize via Posterior Sampling. Mathematics of Operations Research, 39(4), 1221–1243. https://doi.org/10.1287/moor.2014.0650

MLA 9

Russo, Daniel, and Benjamin Van Roy. “Learning to Optimize via Posterior Sampling.” Mathematics of Operations Research, vol. 39, no. 4, 2014, pp. 1221–1243. https://doi.org/10.1287/moor.2014.0650.

Chicago

Russo, Daniel and Benjamin Van Roy. 2014. “Learning to Optimize via Posterior Sampling.” Mathematics of Operations Research 39 (4): 1221–1243. https://doi.org/10.1287/moor.2014.0650.

Harvard

Russo, D. and Van Roy, B. (2014) ‘Learning to Optimize via Posterior Sampling’, Mathematics of Operations Research, 39(4), pp. 1221–1243. doi:10.1287/moor.2014.0650.

IEEE

D. Russo and B. Van Roy, “Learning to Optimize via Posterior Sampling,” Mathematics of Operations Research, vol. 39, no. 4, pp. 1221–1243, 2014, doi: 10.1287/moor.2014.0650.

BibTeX

@article{russo2014learning,
  title     = {{Learning to Optimize via Posterior Sampling}},
  author    = {Russo, Daniel and Van Roy, Benjamin},
  journal   = {Mathematics of Operations Research},
  year      = {2014},
  volume    = {39},
  number    = {4},
  pages     = {1221--1243},
  doi       = {10.1287/moor.2014.0650},
  url       = {https://doi.org/10.1287/moor.2014.0650}
}

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

Citations by year

YearCitationsShare of the best year
20173636
20184141
20195454
2020143143
20218181
20228181
20238787
20247676
20257878
20267979

Its field-weighted citation impact is 27.19: it has been cited 27.19 times as often as the average work of the same field, type and year (1.00 is the world average).

Topic and keywords

Social Sciences › Decision Sciences › Management Science and Operations Research › Advanced Bandit Algorithms Research

  • posterior sampling
  • Thompson sampling
  • multi-armed bandits
  • exploration-exploitation balance
  • upper confidence bound
  • linear bandits
  • generalized linear models
  • regret analysis

Work with this paper

Most-cited papers in Decision Sciences

All highly cited Decision Sciences papers

Questions about this paper

How do I cite this paper in APA style?

Russo, D., & Van Roy, B. (2014). Learning to Optimize via Posterior Sampling. Mathematics of Operations Research, 39(4), 1221–1243. https://doi.org/10.1287/moor.2014.0650

How many times has this paper been cited?

825 times, according to OpenAlex on 19 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 at arXiv (Cornell University): sign in and open the PDF from this page.

Who wrote this paper?

Daniel Russo, Benjamin Van Roy.

What is the DOI of this paper?

10.1287/moor.2014.0650, permanently at https://doi.org/10.1287/moor.2014.0650. Sign in and open it from this page.

Bibliographic data from OpenAlex (CC0), last checked 19 November 2026. ResearchGuru does not host this paper; once you are signed in, the links open it at the publisher or the repository that does.