English

On the Complexity of A/B Testing

Statistics Theory 2015-02-25 v2 Machine Learning Machine Learning Statistics Theory

Abstract

A/B testing refers to the task of determining the best option among two alternatives that yield random outcomes. We provide distribution-dependent lower bounds for the performance of A/B testing that improve over the results currently available both in the fixed-confidence (or delta-PAC) and fixed-budget settings. When the distribution of the outcomes are Gaussian, we prove that the complexity of the fixed-confidence and fixed-budget settings are equivalent, and that uniform sampling of both alternatives is optimal only in the case of equal variances. In the common variance case, we also provide a stopping rule that terminates faster than existing fixed-confidence algorithms. In the case of Bernoulli distributions, we show that the complexity of fixed-budget setting is smaller than that of fixed-confidence setting and that uniform sampling of both alternatives -though not optimal- is advisable in practice when combined with an appropriate stopping criterion.

Keywords

Cite

@article{arxiv.1405.3224,
  title  = {On the Complexity of A/B Testing},
  author = {Emilie Kaufmann and Olivier Cappé and Aurélien Garivier},
  journal= {arXiv preprint arXiv:1405.3224},
  year   = {2015}
}
R2 v1 2026-06-22T04:13:08.628Z