Improving the Estimation of Tail Ratings in Recommender System with Multi-Latent Representations

Zhao, Xing; Zhu, Ziwei; Zhang, Yin; Caverlee, James

doi:10.1145/3336191.3371810

“…Most commonly, a multilayer perceptron (MLP) is used for the network architecture (e.g. [19,21,28,33,38,39]). This approach is often referred to as neural collaborative filtering (NCF) [17].…”

Section: Introductionmentioning

confidence: 99%

“…This has made NCF the model of choice for comparison in many recommender studies (e.g. [19,21,28,31,38,39]).…”

Section: Introductionmentioning

confidence: 99%

Neural Collaborative Filtering vs. Matrix Factorization Revisited

Rendle

¹

,

Krichene

²

,

Zhang

³

et al. 2020

Fourteenth ACM Conference on Recommender Systems

View full text Add to dashboard Cite

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. CCS CONCEPTS • Information systems → Recommender systems; Collaborative filtering; • Computing methodologies → Learning from implicit feedback; Neural networks; Factorization methods.

show abstract

“…Most commonly, a multilayer perceptron (MLP) is used for the network architecture (e.g. [18,37,38,20,32,27]). This approach is often referred to as neural collaborative filtering (NCF) [16].…”

Section: Introductionmentioning

confidence: 99%

“…This has made NCF the model of choice for comparison in many recommender studies (e.g. [18,37,30,38,20,27]).…”

Section: Introductionmentioning

confidence: 99%

Neural Collaborative Filtering vs. Matrix Factorization Revisited

Rendle¹,

Krichene²,

Zhang³

et al. 2020

Preprint

View full text Add to dashboard Cite

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.

show abstract