Neural Collaborative Filtering vs. Matrix Factorization Revisited
This work addresses practical issues in recommendation systems for researchers and practitioners, highlighting that dot products may be a better default choice, though it is incremental as it revisits prior findings.
The authors revisited neural collaborative filtering (NCF) experiments and found that with proper hyperparameter tuning, a simple dot product outperforms learned similarities using MLPs, showing MLPs are costly and non-trivial to learn dot products effectively.
Embedding based models have been the state of the art in collaborative filtering for over a decade. Traditionally, the dot product or higher order equivalents have been used to combine two or more embeddings, e.g., most notably in matrix factorization. In recent years, it was suggested to replace the dot product with a learned similarity e.g. using a multilayer perceptron (MLP). This approach is often referred to as neural collaborative filtering (NCF). In this work, we revisit the experiments of the NCF paper that popularized learned similarities using MLPs. First, we show that with a proper hyperparameter selection, a simple dot product substantially outperforms the proposed learned similarities. Second, while a MLP can in theory approximate any function, we show that it is non-trivial to learn a dot product with an MLP. Finally, we discuss practical issues that arise when applying MLP based similarities and show that MLPs are too costly to use for item recommendation in production environments while dot products allow to apply very efficient retrieval algorithms. We conclude that MLPs should be used with care as embedding combiner and that dot products might be a better default choice.