Combining Reward and Rank Signals for Slate Recommendation
This work addresses the challenge of improving recommendation accuracy in systems that present multiple items at once, though it appears incremental by building on existing signal types.
The paper tackles the problem of slate recommendation by formulating Bayesian models that incorporate reward and rank signals, showing that the combined model achieves significantly lower error as catalog size or slate size increases.
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? (rank). In this paper, we formulate several Bayesian models that incorporate the reward signal (Reward model), the rank signal (Rank model), or both (Full model), 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.