BCIRWIS Workshop at KDD 2022
Combining Reward and Rank Signals for Slate Recommendation
Abstract
We consider the problem of slate recommendation, where the recommender system presents a user with a collection or slate composed of K recommended items at once. If the user finds the recommended items appealing then the user may click and the recommender system receives some feedback. Two pieces of information are available to the recommender system: was the slate clicked, the reward, and if the slate was clicked, which item was clicked, the rank. In this paper, we formulate several Bayesian models that incorporate the reward signal, the rank signal, or both, for non-personalized slate recommendation. In our experiments, we analyze performance gains of the full model and show that it achieves significantly lower error as the number of products in the catalog grows or as the slate size increases.