Things on this page are fragmentary and immature notes/thoughts of the author. Please read with your own judgement!

User-based Filtering (Memory-based Filtering)

Item-based Filtering (Content-based Filtering)

Non-negative Matrix Factorization

Neural Matrix Factorization

Variational Autoencoder

Hybrid

New methods like VAE, AE, or Deep Collaborative outperform classical methods like NMF on the NDCG metric. Non-linear probabilistic models such as variational autoencoders enable us to go beyond the limited modeling capacity of linear factor models.

References

https://medium.com/snipfeed/how-variational-autoencoders-make-classical-recommender-systems-obsolete-4df8bae51546

https://medium.com/snipfeed/how-to-implement-deep-generative-models-for-recommender-systems-29110be8971f

Variational Autoencoders for Collaborative Filtering

https://en.wikipedia.org/wiki/Matrix_factorization_(recommender_systems)