REMIT: Reinforced Multi-Interest Transfer for Cross-Domain Recommendation.

Cold-start problem is one of the most challenging problems for recommender systems. One promising solution to this problem is cross-domain recommendation (CDR) which leverages rich information from an auxiliary source domain to improve the performance of recommender system in the target domain. In particular, the family of embedding and mapping methods for CDR is very effective, which explicitly learns a mapping function from source embeddings to target embeddings to transfer users' preferences. Recent works usually transfer an overall source embedding by modeling a common or personalized preference bridge for all users. However, a unified user embedding cannot reflect the user's multiple interests in the auxiliary source domain. In this paper, we propose a novel framework called RE inforced M ulti I nterest T ransfer for CDR (REMIT). Specifically, we first construct a heterogeneous information network and employ different meta-path based aggregations to get users' multiple interests in the source domain, then transform different interest embeddings with different meta-generated personalized bridge functions for each user. To better coordinate the transformed user interest embeddings and the item embedding in the target domain, we systematically develop a reinforced method to dynamically assign weights to transformed interests for different training instances and optimize the performance of the target model. In addition, the REMIT is a general framework that can be applied upon various base models in the target domain. Our extensive experimental results on large real-world datasets demonstrate the superior performance and compatibility of REMIT.