Combines the benefits of memorization (wide linear models) and generalization (deep neural networks).
Describes a two-stage system consisting of Candidate Generation and Ranking . 117371782_294128495019917_5947729689354735659_n...
One of the most cited industrial deep learning papers, explaining how YouTube uses deep learning to process billions of videos and user interactions. 3. Deep Learning-Based Recommendation: A Survey Authors: Various (e.g., Zhang et al.) Combines the benefits of memorization (wide linear models)
Provides a comprehensive taxonomy of deep learning models used in recommendation, such as CNNs, RNNs, and Restricted Boltzmann Machines (RBMs). such as CNNs
The most prominent "deep papers" in the recommendation domain include: 1. Authors: Cheng et al. (Google)
