On Multi-Armed Bandit Designs for Dose-Finding Trials - Inria - Institut national de recherche en sciences et technologies du numérique Access content directly
Journal Articles Journal of Machine Learning Research Year : 2021

On Multi-Armed Bandit Designs for Dose-Finding Trials

Maryam Aziz
  • Function : Author
Emilie Kaufmann
Marie-Karelle Riviere
  • Function : Author
  • PersonId : 980526


We study the problem of finding the optimal dosage in early stage clinical trials through the multi-armed bandit lens. We advocate the use of the Thompson Sampling principle, a flexible algorithm that can accommodate different types of monotonicity assumptions on the toxicity and efficacy of the doses. For the simplest version of Thompson Sampling, based on a uniform prior distribution for each dose, we provide finite-time upper bounds on the number of sub-optimal dose selections, which is unprecedented for dose-finding algorithms. Through a large simulation study, we then show that variants of Thompson Sampling based on more sophisticated prior distributions outperform state-of-the-art dose identification algorithms in different types of dose-finding studies that occur in phase I or phase I/II trials.
Fichier principal
Vignette du fichier
AKR_ClinicalTrials20.pdf (445.42 Ko) Télécharger le fichier
doses.pdf (10.98 Ko) Télécharger le fichier
doses2.pdf (10.97 Ko) Télécharger le fichier
illustration.pdf (8.86 Ko) Télécharger le fichier
Origin : Files produced by the author(s)
Origin : Files produced by the author(s)

Dates and versions

hal-02533297 , version 1 (06-04-2020)


  • HAL Id : hal-02533297 , version 1


Maryam Aziz, Emilie Kaufmann, Marie-Karelle Riviere. On Multi-Armed Bandit Designs for Dose-Finding Trials. Journal of Machine Learning Research, 2021. ⟨hal-02533297⟩
277 View
361 Download


Gmail Facebook X LinkedIn More