Pessimistic Reward Models for Off-Policy Learning in Recommendation.

Olivier Jeunen, Bart Goethals

Browse the full RecSys paper archive.