On Multi-Armed Bandit Designs for Dose-Finding ...
Document type :
Compte-rendu et recension critique d'ouvrage
Title :
On Multi-Armed Bandit Designs for Dose-Finding Trials
Author(s) :
Aziz, Maryam [Auteur]
SPOTIFY [London]
Kaufmann, Emilie [Auteur]
Scool [Scool]
Riviere, Marie-Karelle [Auteur]
SANOFI Recherche
SPOTIFY [London]
Kaufmann, Emilie [Auteur]
Scool [Scool]
Riviere, Marie-Karelle [Auteur]
SANOFI Recherche
Journal title :
Journal of Machine Learning Research
Publisher :
Microtome Publishing
Publication date :
2021-01
ISSN :
1532-4435
HAL domain(s) :
Statistiques [stat]/Machine Learning [stat.ML]
English abstract : [en]
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 ...
Show more >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.Show less >
Show more >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.Show less >
Language :
Anglais
Popular science :
Non
ANR Project :
Collections :
Source :
Files
- https://hal.archives-ouvertes.fr/hal-02533297/document
- Open access
- Access the document
- https://hal.archives-ouvertes.fr/hal-02533297/file/doses.pdf
- Open access
- Access the document
- https://hal.archives-ouvertes.fr/hal-02533297/file/doses2.pdf
- Open access
- Access the document
- https://hal.archives-ouvertes.fr/hal-02533297/file/illustration.pdf
- Open access
- Access the document
- https://hal.archives-ouvertes.fr/hal-02533297/document
- Open access
- Access the document
- https://hal.archives-ouvertes.fr/hal-02533297/document
- Open access
- Access the document
- AKR_ClinicalTrials20.pdf
- Open access
- Access the document
- doses.pdf
- Open access
- Access the document
- illustration.pdf
- Open access
- Access the document
- document
- Open access
- Access the document
- doses2.pdf
- Open access
- Access the document