• Corpus ID: 159042202

RaFM: Rank-Aware Factorization Machines

@inproceedings{Chen2019RaFMRF,
  title={RaFM: Rank-Aware Factorization Machines},
  author={Xiaoshuang Chen and Yin Zheng and Jiaxing Wang and Wenye Ma and Junzhou Huang},
  booktitle={ICML},
  year={2019}
}
Factorization machines (FM) are a popular model class to learn pairwise interactions by a low-rank approximation. Different from existing FM-based approaches which use a fixed rank for all features, this paper proposes a Rank-Aware FM (RaFM) model which adopts pairwise interactions from embeddings with different ranks. The proposed model achieves a better performance on real-world datasets where different features have significantly varying frequencies of occurrences. Moreover, we prove that… 

Figures and Tables from this paper

AdnFM: An Attentive DenseNet based Factorization Machine for Click-Through-Rate Prediction
TLDR
A novel deep learning-based model named AdnFM, to attack the Click-Through-Rate (CTR) prediction problem, which combines residual learning and an attention mechanism, to enable high-order features interactions and weight their importance dynamically.
AdnFM: An Attentive DenseNet based Factorization Machine for CTR Prediction
TLDR
A novel deep learning-based model named AdnFM, to attack the Click-Through-Rate (CTR) prediction problem, which combines residual learning and an attention mechanism, to enable high-order features interactions and weight their importance dynamically.
Deep Interest-Shifting Network with Meta-Embeddings for Fresh Item Recommendation
TLDR
A deep interest-shifting network (DisNet), which transfers knowledge from a huge number of auxiliary data and then shifts user interests with contextual information and a relational meta-Id-embedding generator (RM-IdEG) that integrates relational information for better embedding performance is proposed.
Factorization Machines with Regularization for Sparse Feature Interactions
TLDR
A new regularization scheme for feature interaction selection in FMs is presented and an upper bound of the $\ell_1$ regularizer for the feature interaction matrix is computed from the parameter matrix of FMs.
Field-wise Learning for Multi-field Categorical Data
TLDR
This work presents a model that utilizes linear models with variance and low-rank constraints, to help it generalize better and reduce the number of parameters, and is also interpretable in a field-wise manner.
Projective Quadratic Regression for Online Learning
TLDR
A projective quadratic regression (PQR) model is proposed that can capture the import second-order feature information and is a convex model, so the requirements of OCO are fulfilled and the global optimal solution can be achieved.

References

SHOWING 1-10 OF 18 REFERENCES
A Boosting Framework of Factorization Machine
TLDR
This work proposes an Adaptive Boosting framework of Factorization Machines (AdaFM), which can adaptively search for proper ranks for different datasets without re-training, and will adaptively gradually increases its rank according to its performance until the performance does not grow, using boosting strategy.
Field-aware Factorization Machines for CTR Prediction
TLDR
This paper establishes FFMs as an effective method for classifying large sparse data including those from CTR prediction, and proposes efficient implementations for training FFMs and comprehensively analyze FFMs.
Locally Linear Factorization Machines
TLDR
This work forms a joint optimization over the anchor points, local coding coordinates and FMs variables to minimize classification or regression risk and demonstrates that this approach achieves much better predictive accuracy than other competitive methods which employ LLFM with unsupervised anchor point learning and predefined local coding scheme.
Convex Factorization Machines
TLDR
This paper proposes a convex formulation of factorization machines based on the nuclear norm, which imposes fewer restrictions on the learned model and is thus more general than the original formulation and presents an efficient globally-convergent two-block coordinate descent algorithm.
Attentional Factorization Machines: Learning the Weight of Feature Interactions via Attention Networks
TLDR
A novel model named Attentional Factorization Machine (AFM), which learns the importance of each feature interaction from data via a neural attention network, which consistently outperforms the state-of-the-art deep learning methods Wide&Deep and DeepCross with a much simpler structure and fewer model parameters.
Factorization Machines with libFM
Factorization approaches provide high accuracy in several important prediction problems, for example, recommender systems. However, applying factorization approaches to a new prediction problem is a
DeepFM: A Factorization-Machine based Neural Network for CTR Prediction
TLDR
This paper shows that it is possible to derive an end-to-end learning model that emphasizes both low- and high-order feature interactions, and combines the power of factorization machines for recommendation and deep learning for feature learning in a new neural network architecture.
Mixture-Rank Matrix Approximation for Collaborative Filtering
TLDR
A mixture-rank matrix approximation (MRMA) method is proposed, in which user-item ratings can be characterized by a mixture of LRMA models with different ranks, and a learning algorithm capitalizing on iterated condition modes is proposed to tackle the non-convex optimization problem pertaining to MRMA.
Factorization Machines
  • Steffen Rendle
  • Computer Science
    2010 IEEE International Conference on Data Mining
  • 2010
TLDR
Factorization Machines (FM) are introduced which are a new model class that combines the advantages of Support Vector Machines (SVM) with factorization models and can mimic these models just by specifying the input data (i.e. the feature vectors).
Neural Factorization Machines for Sparse Predictive Analytics
TLDR
NFM seamlessly combines the linearity of FM in modelling second- order feature interactions and the non-linearity of neural network in modelling higher-order feature interactions, and is more expressive than FM since FM can be seen as a special case of NFM without hidden layers.
...
...