A Neural Autoregressive Approach to Collaborative Filtering

TL;DR

This paper introduces CF-NADE, a neural autoregressive model for collaborative filtering that outperforms previous methods on major datasets, with improvements through parameter sharing, scalability, and deep architectures.

Contribution

The paper presents a novel neural autoregressive architecture for collaborative filtering, incorporating parameter sharing, scalability, ordinal preferences, and deep models, achieving state-of-the-art results.

Findings

CF-NADE outperforms previous methods on MovieLens and Netflix datasets.

Parameter sharing and ordinal cost improve model performance.

Deeper CF-NADE models further enhance recommendation accuracy.

Abstract

This paper proposes CF-NADE, a neural autoregressive architecture for collaborative filtering (CF) tasks, which is inspired by the Restricted Boltzmann Machine (RBM) based CF model and the Neural Autoregressive Distribution Estimator (NADE). We first describe the basic CF-NADE model for CF tasks. Then we propose to improve the model by sharing parameters between different ratings. A factored version of CF-NADE is also proposed for better scalability. Furthermore, we take the ordinal nature of the preferences into consideration and propose an ordinal cost to optimize CF-NADE, which shows superior performance. Finally, CF-NADE can be extended to a deep model, with only moderately increased computational complexity. Experimental results show that CF-NADE with a single hidden layer beats all previous state-of-the-art methods on MovieLens 1M, MovieLens 10M, and Netflix datasets, and adding more hidden layers can further improve the performance.