Can I read Deep Autoencoders for Collaborative Filtering on EtoBox?

Deep Autoencoders for Collaborative Filtering by ĐứcĐiệu is a document available to read on EtoBox.

What is Deep Autoencoders for Collaborative Filtering about?

Deep autoencoders can be used for collaborative filtering to make personalized recommendations. The paper proposes a model that significantly outperforms previous methods on a Netflix dataset. It uses a deep autoencoder with 6 layers trained end-to-end without pre-training. Experimental results show that deep autoencoders generalize better than shallow ones, negative activation functions are crucial for training deep models, and regularization like dropout is needed to prevent overfitting.

Author

ĐứcĐiệu

Language

EN