In this article, we explain why an interventional view of recommendation provides a rigorous framework for thinking about recommender systems—enabling new insights both at a technical level for evaluation and learning, as well as at a conceptual level when we reason about the future of recommender systems. In some respects, the view of recommender systems as autonomous systems that act through their recommendations is already part of common industry practice. For example, A/B tests are widely recognized as the gold standard in evaluating recommender systems, and they are functionally equivalent to a controlled randomized trial in medicine. However, we argue that the connection between recommender systems and causal inference runs much deeper, leading to a rigorous foundation for the field that produces new algorithms with provable guarantees.
Research areas
