arXiv 16 Feb 2024 · Machine Learning · 3 citations (OpenAlex)
arXiv:2402.10592 · PDF · DOI · OpenAlex · Extracted main text
Practitioners conducting adaptive experiments often encounter two competing priorities: maximizing total welfare (or `reward') through effective treatment assignment and swiftly concluding experiments to implement population-wide treatments. Current literature addresses these priorities separately, with regret minimization studies focusing on the former and best-arm identification research on the latter. This paper bridges this divide by proposing a unified model that simultaneously accounts for within-experiment performance and post-experiment outcomes. We provide a sharp theory of optimal performance in large populations that not only unifies canonical results in the literature but also uncovers novel insights. Our theory reveals that familiar algorithms, such as the recently proposed top-two Thompson sampling algorithm, can optimize a broad class of objectives if a single scalar parameter is appropriately adjusted. In addition, we demonstrate that substantial reductions in experiment duration can often be achieved with minimal impact on both within-experiment and post-experiment regret.
appendix boundary found by appendix_command · 38% of the source is main text. Read the extracted text to check this.
The works this paper leans on most, across its whole bibliography — not restricted to papers in our corpus. Ranked by composite intensity, which combines how often a work is mentioned, how many sections mention it, and how much of that falls in the main text rather than the appendix.
| Reference | Intensity | Mentions | Sections | Main text | |
|---|---|---|---|---|---|
| 1 | Tze Leung Lai and Herbert Robbins (1985) Asymptotically efficient adaptive allocation rules | 1.000 | 16 | 6 | 100% |
| 2 | Daniel Russo (2016) Simple bayesian algorithms for best arm identification self | 1.000 | 5 | 4 | 100% |
| 3 | Daniel Russo (2020) Simple bayesian algorithms for best-arm identification self | 0.950 | 14 | 6 | 86% |
| 4 | Herman Chernoff (1959) Sequential design of experiments | 0.941 | 6 | 5 | 83% |
| 5 | Hock Peng Chan and Tze Leung Lai (2006) Sequential generalized likelihood ratios and adaptive treatment allocation for optimal sequential selection | 0.874 | 5 | 2 | 100% |
| 6 | Emilie Kaufmann, Olivier Cappé, and Aurélien Garivier (2016) On the complexity of best-arm identification in multi-armed bandit models | 0.843 | 3 | 3 | 100% |
| 7 | Peter Glynn and Sandeep Juneja (2004) A large deviations perspective on ordinal optimization | 0.811 | 4 | 2 | 100% |
| 8 | Tor Lattimore and Csaba Szepesvári (2020) Bandit algorithms | 0.794 | 6 | 5 | 50% |
| 9 | Chao Qin, Diego Klabjan, and Daniel Russo (2017) Improving the expected improvement algorithm self | 0.737 | 5 | 3 | 40% |
| 10 | Aurélien Garivier and Emilie Kaufmann (2016) Optimal best arm identification with fixed confidence | 0.716 | 30 | 10 | 37% |
Showing the top 10 of 72 scored citations.
arXiv econ.EM papers that cite this one, ranked by how heavily they lean on it.
| Citing paper | Intensity | Mentions | Sections | |
|---|---|---|---|---|
| 1 | Bandit Algorithms for Policy Learning: Methods, Implementation, and Welfare-performance | 0.405 | 1 | 1 |
| 2 | Admissibility of Completely Randomized Trials: A Large-Deviation Approach | 0.405 | 1 | 1 |