Negative Interactions for Improved Collaborative-Filtering: Don’t go Deeper, go Higher скачать в хорошем качестве

Negative Interactions for Improved Collaborative-Filtering: Don’t go Deeper, go Higher

RecSys 2021 Negative Interactions for Improved Collaborative-Filtering: Don’t go Deeper, go Higher Authors: Harald Steck, Netflix | Dawen Liang, Netflix, Inc. Abstract:The recommendation-accuracy of collaborative-filtering approaches is typically improved when taking into account higher-order interactions. While deep nonlinear models are theoretically able to learn higher-order interactions, their capabilities were, however, found to be quite limited in practice. Moreover, the use of low-dimensional embeddings in deep networks may severely limit their expressiveness. This motivated us in this paper to explore a simple extension of linear full-rank models that allow for higher-order interactions as additional explicit input-features. Interestingly, we observed that this model-class obtained by far the best ranking-accuracies on the largest data set in our experiments, while it was still competitive with various state-of-the-art deep-learning models on the smaller data sets. Moreover, our approach can also be interpreted as a simple yet effective improvement of the (linear) HOSLIM model: by simply removing the constraint that the learned higher-order interactions have to be non-negative, we observed that the accuracy-gains due to higher-order interactions more than doubled in our experiments. The reason for this large improvement was that large positive higher-order interactions (as used in HOSLIM) are relatively infrequent compared to the number of large negative higher-order interactions in the three well-known data-sets used in our experiments. We further characterize the circumstances where the higher-order interactions provide the most significant improvements. DOI: https://doi.org/10.1145/3460231.3474273