One-class Field-aware Factorization Machines for Recommender Systems with Implicit Feedbacks

Recommender systems with implicit feedbacks is a typical one-class scenario, where only positive labels are available. Such positive-unlabeled (PU) learning problems can be solved by one-class matrix factorization (OCMF). Recently, OCMF with side information (OCMFSI) on users and items has been proposed as a powerful extension of OCMF. Interestingly, OCMFSI is strongly related to Factorization Machines (FM), which is a general classification and regression model. This link motivates us to investigate if models superior to FM (e.g., Field-aware Factorization Machines) can be effectively extended to PU learning. In this paper, we propose a novel One-class Field-aware Factorization Machines (OCFFM) model. An efficient optimization algorithm is developed such that OCFFM can be trained on the large-scale data sets. Finally, through experiments on four data sets, OCFFM shows its superiority over other one-class models.

[1]  M. Hestenes,et al.  Methods of conjugate gradients for solving linear systems , 1952 .

[2]  Chih-Jen Lin,et al.  Trust region Newton methods for large-scale logistic regression , 2007, ICML '07.

[3]  Qiang Yang,et al.  One-Class Collaborative Filtering , 2008, 2008 Eighth IEEE International Conference on Data Mining.

[4]  Yifan Hu,et al.  Collaborative Filtering for Implicit Feedback Datasets , 2008, 2008 Eighth IEEE International Conference on Data Mining.

[5]  Lars Schmidt-Thieme,et al.  BPR: Bayesian Personalized Ranking from Implicit Feedback , 2009, UAI.

[6]  Rong Pan,et al.  Mind the gaps: weighting the unknown in large-scale one-class collaborative filtering , 2009, KDD.

[7]  Steffen Rendle,et al.  Factorization Machines , 2010, 2010 IEEE International Conference on Data Mining.

[8]  Lars Schmidt-Thieme,et al.  Learning Attribute-to-Feature Mappings for Cold-Start Recommendations , 2010, 2010 IEEE International Conference on Data Mining.

[9]  Chia-Hua Ho,et al.  An improved GLMNET for l1-regularized logistic regression , 2011, J. Mach. Learn. Res..

[10]  Steffen Rendle,et al.  Factorization Machines with libFM , 2012, TIST.

[11]  Jun Wang,et al.  Optimizing top-n collaborative filtering via dynamic negative item sampling , 2013, SIGIR.

[12]  Ulrich Paquet,et al.  One-class collaborative filtering with random graphs , 2013, WWW '13.

[13]  Tong Zhao,et al.  Leveraging Social Connections to Improve Personalized Ranking for Collaborative Filtering , 2014, CIKM.

[14]  Nitish Srivastava,et al.  Dropout: a simple way to prevent neural networks from overfitting , 2014, J. Mach. Learn. Res..

[15]  Chih-Jen Lin,et al.  Fast Matrix-Vector Multiplications for Large-Scale Logistic Regression on Shared-Memory Systems , 2015, 2015 IEEE International Conference on Data Mining.

[16]  Jimmy Ba,et al.  Adam: A Method for Stochastic Optimization , 2014, ICLR.

[17]  Nagarajan Natarajan,et al.  PU Learning for Matrix Completion , 2014, ICML.

[18]  Naonori Ueda,et al.  Polynomial Networks and Factorization Machines: New Insights and Efficient Training Algorithms , 2016, ICML.

[19]  Chih-Jen Lin,et al.  Field-aware Factorization Machines for CTR Prediction , 2016, RecSys.

[20]  Tat-Seng Chua,et al.  Fast Matrix Factorization for Online Recommendation with Implicit Feedback , 2016, SIGIR.

[21]  Yunming Ye,et al.  DeepFM: A Factorization-Machine based Neural Network for CTR Prediction , 2017, IJCAI.

[22]  Xiangnan He,et al.  A Generic Coordinate Descent Framework for Learning from Implicit Feedback , 2016, WWW.

[23]  Chih-Jen Lin,et al.  Selection of Negative Samples for One-class Matrix Factorization , 2017, SDM.

[24]  Chih-Jen Lin,et al.  A Unified Algorithm for One-Cass Structured Matrix Factorization with Side Information , 2017, AAAI.

[25]  Olivier Chapelle,et al.  Field-aware Factorization Machines in a Real-world Online Advertising System , 2017, WWW.

[26]  Guy Blanc,et al.  Adaptive Sampled Softmax with Kernel Based Sampling , 2017, ICML.

[27]  Xing Zhang,et al.  CPLR: Collaborative pairwise learning to rank for personalized recommendation , 2018, Knowl. Based Syst..

[28]  Chih-Jen Lin,et al.  An Efficient Alternating Newton Method for Learning Factorization Machines , 2018, ACM Trans. Intell. Syst. Technol..

[29]  Guorui Zhou,et al.  Deep Interest Network for Click-Through Rate Prediction , 2017, KDD.

[30]  Yun Liu,et al.  BPRH: Bayesian personalized ranking for heterogeneous implicit feedback , 2018, Inf. Sci..

[31]  Tat-Seng Chua,et al.  fBGD: Learning Embeddings From Positive Unlabeled Data with BGD , 2018, UAI.

[32]  Li Chen,et al.  Neighborhood-enhanced transfer learning for one-class collaborative filtering , 2019, Neurocomputing.

[33]  Natalia Gimelshein,et al.  PyTorch: An Imperative Style, High-Performance Deep Learning Library , 2019, NeurIPS.

[34]  Kai Zheng,et al.  Improving One-Class Collaborative Filtering via Ranking-Based Implicit Regularizer , 2019, AAAI.

[35]  John R. Anderson,et al.  Efficient Training on Very Large Corpora via Gramian Estimation , 2018, ICLR.

[36]  Wang Zhou,et al.  Bayesian pairwise learning to rank via one-class collaborative filtering , 2019, Neurocomputing.

[37]  Yiqun Liu,et al.  Efficient Neural Matrix Factorization without Sampling for Recommendation , 2020, ACM Trans. Inf. Syst..

[38]  Enhong Chen,et al.  Personalized Ranking with Importance Sampling , 2020, WWW.

[39]  Yiqun Liu,et al.  Efficient Non-Sampling Factorization Machines for Optimal Context-Aware Recommendation , 2020, WWW.