Attentional Factorization Machines: Learning the Weight of Feature Interactions via Attention Networks
This work addresses a bottleneck in recommendation systems and predictive modeling by improving feature interaction weighting, though it is incremental as it builds directly on FM.
The paper tackles the problem of Factorization Machines (FMs) assigning equal weight to all feature interactions, which can degrade performance due to noise from useless features, by proposing Attentional Factorization Machine (AFM) that learns importance via a neural attention network, achieving an 8.6% relative improvement over FM on regression tasks and outperforming state-of-the-art deep learning methods with simpler structure and fewer parameters.
Factorization Machines (FMs) are a supervised learning approach that enhances the linear regression model by incorporating the second-order feature interactions. Despite effectiveness, FM can be hindered by its modelling of all feature interactions with the same weight, as not all feature interactions are equally useful and predictive. For example, the interactions with useless features may even introduce noises and adversely degrade the performance. In this work, we improve FM by discriminating the importance of different feature interactions. We propose a novel model named Attentional Factorization Machine (AFM), which learns the importance of each feature interaction from data via a neural attention network. Extensive experiments on two real-world datasets demonstrate the effectiveness of AFM. Empirically, it is shown on regression task AFM betters FM with a $8.6\%$ relative improvement, and consistently outperforms the state-of-the-art deep learning methods Wide&Deep and DeepCross with a much simpler structure and fewer model parameters. Our implementation of AFM is publicly available at: https://github.com/hexiangnan/attentional_factorization_machine