Learning to Recommend from Sparse Data via Generative User Feedback

Traditional collaborative filtering (CF) based recommender systems tend to perform poorly when the user-item interactions/ratings are highly scarce. To address this, we propose a learning framework that improves with synthetic feedback loop (CF-SFL) simulate user feedback. The proposed consists of and virtual user. is formulated as CF model, recommending items according observed preference. estimates rewards from recommended generates in addition connected constructs closed loop, recommends users imitates unobserved items. used augment preference improve recommendation results. Theoretically, such model design can be interpreted inverse reinforcement learning, which learned effectively via rollout (simulation). Experimental results show able enrich boost performance existing methods on multiple datasets.