Stochastic Bandit Models for Delayed Conversions

Abstract : Online advertising and product recommendation are important domains of applications for multi-armed bandit methods. In these fields, the reward that is immediately available is most often only a proxy for the actual outcome of interest, which we refer to as a conversion. For instance, in web advertising, clicks can be observed within a few seconds after an ad display but the corresponding sale –if any– will take hours, if not days to happen. This paper proposes and investigates a new stochas-tic multi-armed bandit model in the framework proposed by Chapelle (2014) –based on empirical studies in the field of web advertising– in which each action may trigger a future reward that will then happen with a stochas-tic delay. We assume that the probability of conversion associated with each action is unknown while the distribution of the conversion delay is known, distinguishing between the (idealized) case where the conversion events may be observed whatever their delay and the more realistic setting in which late conversions are censored. We provide performance lower bounds as well as two simple but efficient algorithms based on the UCB and KLUCB frameworks. The latter algorithm, which is preferable when conversion rates are low, is based on a Poissonization argument, of independent interest in other settings where aggregation of Bernoulli observations with different success probabilities is required.
Type de document :
Communication dans un congrès
Conference on Uncertainty in Artificial Intelligence, Aug 2017, Sydney, Australia. 〈http://auai.org/uai2017/index.php〉
Liste complète des métadonnées

https://hal.archives-ouvertes.fr/hal-01545667
Contributeur : Claire Vernade <>
Soumis le : mardi 4 juillet 2017 - 00:59:33
Dernière modification le : jeudi 11 janvier 2018 - 06:27:34
Document(s) archivé(s) le : mardi 23 janvier 2018 - 18:01:11

Fichiers

Paper.pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

  • HAL Id : hal-01545667, version 2
  • ARXIV : 1706.09186

Citation

Claire Vernade, Olivier Cappé, Vianney Perchet. Stochastic Bandit Models for Delayed Conversions. Conference on Uncertainty in Artificial Intelligence, Aug 2017, Sydney, Australia. 〈http://auai.org/uai2017/index.php〉. 〈hal-01545667v2〉

Partager

Métriques

Consultations de la notice

197

Téléchargements de fichiers

80