Seq2Slate: Re-ranking and Slate Optimization with RNNs
This work addresses the challenge of optimizing slates for users in ranking and recommendation systems, offering a scalable method that captures item dependencies, though it is incremental as it builds on existing sequence-to-sequence and ranking techniques.
The paper tackles the problem of ranking items into a slate by considering item interactions, proposing Seq2Slate, a sequence-to-sequence model that predicts the next best item sequentially. It demonstrates effectiveness through experiments on benchmarks and a real-world recommendation system, achieving improved performance with end-to-end learning from click-through data.
Ranking is a central task in machine learning and information retrieval. In this task, it is especially important to present the user with a slate of items that is appealing as a whole. This in turn requires taking into account interactions between items, since intuitively, placing an item on the slate affects the decision of which other items should be placed alongside it. In this work, we propose a sequence-to-sequence model for ranking called seq2slate. At each step, the model predicts the next `best' item to place on the slate given the items already selected. The sequential nature of the model allows complex dependencies between the items to be captured directly in a flexible and scalable way. We show how to learn the model end-to-end from weak supervision in the form of easily obtained click-through data. We further demonstrate the usefulness of our approach in experiments on standard ranking benchmarks as well as in a real-world recommendation system.