Abstract:Modeling feature interactions is of crucial importance to predict click-through rate (CTR) in industrial recommender systems. However, manually crafting cross features usually requires extensive domain knowledge and labor-intensive feature engineering to obtain the desired cross features. To alleviate this problem, the factorization machine (FM) is proposed to model feature interactions from raw features automatically. In particular, it embeds each feature in a vector representation and discovers second-order interactions as the product of two feature representations. In order to learn nonlinear and complex patterns, recent works, such as NFM, PIN, and DeepFM, exploited deep learning techniques to capture higher-order feature interactions. These approaches lack guarantees about the effectiveness of high-order pattern as they model feature interactions in a rather implicit way. To address this limitation, xDeepFM is recently proposed to generate high-order interactions of features in an explicit fashion, where multiple interaction networks are stacked. Nevertheless, xDeepFM suffers from rather high complexity which easily leads to overfitting.In this paper, we develop a more expressive but lightweight solution based on FM, named High-order Attentive Factorization Machine (HoAFM), by accounting for the higher-order sparse feature interactions in an explicit manner. Beyond the linearity of FM, we devise a cross interaction layer, which updates a feature’s representation by aggregating the representations of other co-occurred features. In addition, we perform a bit-wise attention mechanism to determine the different importance of co-occurred features on the granularity of dimensions. By stacking multiple cross interaction layers, we can inject high-order feature interactions into feature representation learning, in order to establish expressive and informative cross features. Extensive experiments are performed on two benchmark datasets, Criteo and Avazu, to demonstrate the rationality and effectiveness of HoAFM. Empirical results suggest that HoAFM achieves significant improvement over other state-of-the-art methods, such as NFM and xDeepFM. https://github.com/zltao/HoAFM.

An Intelligent Field-Aware Factorization Machine Model

L G ] 1 7 A pr 2 01 8 A Boosting Framework of Factorization Machine

A Boosting Framework of Factorization Machine.

Attention-over-Attention Field-Aware Factorization Machine.

Strongly Hierarchical Factorization Machines and ANOVA Kernel Regression

Interaction-aware Factorization Machines for Recommender Systems

F 2 M : Scalable Field-Aware Factorization Machines

Improving Recommendations With Collaborative Factors

An Input-aware Factorization Machine for Sparse Prediction.

Extending the Fama-French Five-Facor Model with the Long Memory Factor Compared with Machine Learning

Enhancing Factorization Machines with Generalized Metric Learning

Robust Factorization Machine - A Doubly Capped Norms Minimization.

DGFFM: Generalized Field-aware Factorization Machine Based on DenseNet.

Fast Convolutional Factorization Machine with Enhanced Robustness

Field-Aware Neural Factorization Machine for Click-Through Rate Prediction

Discrete Factorization Machines for Fast Feature-based Recommendation

Attentional Factorization Machines: Learning the Weight of Feature Interactions via Attention Networks

DexDeepFM: Ensemble Diversity Enhanced Extreme Deep Factorization Machine Model

HoAFM: A High-order Attentive Factorization Machine for CTR Prediction

Field-aware Factorization Machines in a Real-world Online Advertising System