Timezone: »

Sequential Counterfactual Risk Minimization
Houssam Zenati · Eustache Diemert · Matthieu Martin · Julien Mairal · Pierre Gaillard

Wed Jul 26 02:00 PM -- 03:30 PM (PDT) @ Exhibit Hall 1 #311

Counterfactual Risk Minimization (CRM) is a framework for dealing with the logged bandit feedback problem, where the goal is to improve a logging policy using offline data. In this paper, we explore the case where it is possible to deploy learned policies multiple times and acquire new data. We extend the CRM principle and its theory to this scenario, which we call "Sequential Counterfactual Risk Minimization (SCRM)." We introduce a novel counterfactual estimator and identify conditions that can improve the performance of CRM in terms of excess risk and regret rates, by using an analysis similar to restart strategies in accelerated optimization methods. We also provide an empirical evaluation of our method in both discrete and continuous action settings, and demonstrate the benefits of multiple deployments of CRM.

Author Information

Houssam Zenati (Criteo, INRIA)
Eustache Diemert (Criteo AI Lab)
Matthieu Martin (Swiss Federal Institute of Technology Lausanne)
Julien Mairal (Inria)
Pierre Gaillard (INRIA)

More from the Same Authors