Multi-Selection for Recommendation Systems
By: Sahasrajit Sarmasarkar , Zhihao Jiang , Ashish Goel and more
Potential Business Impact:
Keeps your movie picks private, still good.
We present the construction of a multi-selection model to answer differentially private queries in the context of recommendation systems. The server sends back multiple recommendations and a ``local model'' to the user, which the user can run locally on its device to select the item that best fits its private features. We study a setup where the server uses a deep neural network (trained on the Movielens 25M dataset as the ground truth for movie recommendation. In the multi-selection paradigm, the average recommendation utility is approximately 97\% of the optimal utility (as determined by the ground truth neural network) while maintaining a local differential privacy guarantee with $\epsilon$ ranging around 1 with respect to feature vectors of neighboring users. This is in comparison to an average recommendation utility of 91\% in the non-multi-selection regime under the same constraints.
Similar Papers
Privacy-Utility-Bias Trade-offs for Privacy-Preserving Recommender Systems
Machine Learning (CS)
Protects user data while still suggesting good movies.
Research on the Design of a Short Video Recommendation System Based on Multimodal Information and Differential Privacy
Information Retrieval
Keeps your video likes private while showing you good videos.
MUSS: Multilevel Subset Selection for Relevance and Diversity
Machine Learning (CS)
Finds the best mix of items, much faster.