Causal Embeddings for Recommendation: An Extended Abstract
This addresses a fundamental gap in recommendation systems for businesses seeking to directly optimize commercial outcomes rather than user engagement metrics.
The paper tackles the problem that recommendation systems are typically optimized to match past user behavior rather than achieve business objectives like increasing sales. They propose a new learning setup that optimizes for the Incremental Treatment Effect, showing significant improvements over state-of-the-art factorization methods and causal recommendation approaches.
Recommendations are commonly used to modify user's natural behavior, for example, increasing product sales or the time spent on a website. This results in a gap between the ultimate business objective and the classical setup where recommendations are optimized to be coherent with past user behavior. To bridge this gap, we propose a new learning setup for recommendation that optimizes for the Incremental Treatment Effect (ITE) of the policy. We show this is equivalent to learning to predict recommendation outcomes under a fully random recommendation policy and propose a new domain adaptation algorithm that learns from logged data containing outcomes from a biased recommendation policy and predicts recommendation outcomes according to random exposure. We compare our method against state-of-the-art factorization methods, in addition to new approaches of causal recommendation and show significant improvements.