Improved Bounded Matrix Completion for Large-Scale Recommender Systems

Matrix completion is a widely used technique for personalized recommender systems. In this paper, we focus on the idea of Bounded Matrix Completion (BMC) which imposes bounded constraints into the standard matrix completion problem. It has been shown that BMC works well for several real world datasets, and an efficient coordinate descent solver called BMA has been proposed in [R. Kannan and Park., 2012]. However, we observe that BMA can sometimes converge to nonstationary points, resulting in a relatively poor accuracy in those cases. To overcome this issue, we propose our new approach for solving BMC under the ADMM framework. The proposed algorithm is guaranteed to converge to stationary points. Experimental results on real world datasets show that our algorithm can reach a lower objective function value, obtain a higher prediction accuracy and have better scalability compared with existing bounded matrix completion approaches. Moreover, our method outperforms the state-of-art standard matrix factorization in terms of prediction error in many real datasets.

[1]  Michael I. Jordan,et al.  Advances in Neural Information Processing Systems 30 , 1995 .

[2]  Inderjit S. Dhillon,et al.  NOMAD: Nonlocking, stOchastic Multi-machine algorithm for Asynchronous and Decentralized matrix completion , 2013, Proc. VLDB Endow..

[3]  Haesun Park,et al.  Bounded Matrix Low Rank Approximation , 2012, 2012 IEEE 12th International Conference on Data Mining.

[4]  Emmanuel J. Candès,et al.  Exact Matrix Completion via Convex Optimization , 2009, Found. Comput. Math..

[5]  Dinh Phung,et al.  Journal of Machine Learning Research: Preface , 2014 .

[6]  Nikhil S. Ketkar.  Stochastic Gradient Descent , 2017 .

[7]  Stephen P. Boyd,et al.  Distributed Optimization and Statistical Learning via the Alternating Direction Method of Multipliers , 2011, Found. Trends Mach. Learn..

[8]  Yehuda Koren,et al.  Factorization meets the neighborhood: a multifaceted collaborative filtering model , 2008, KDD.

[9]  Mu Zhu,et al.  Content‐boosted matrix factorization techniques for recommender systems , 2012, Stat. Anal. Data Min..

[10]  Inderjit S. Dhillon,et al.  Matrix Completion with Noisy Side Information , 2015, NIPS.

[11]  Robert Tibshirani,et al.  Spectral Regularization Algorithms for Learning Large Incomplete Matrices , 2010, J. Mach. Learn. Res..

[12]  Nathan Halko,et al.  Finding Structure with Randomness: Probabilistic Algorithms for Constructing Approximate Matrix Decompositions , 2009, SIAM Rev..

[13]  R. Larsen.  Lanczos Bidiagonalization With Partial Reorthogonalization , 1998 .

[14]  Zhixun Su,et al.  Linearized Alternating Direction Method with Adaptive Penalty for Low-Rank Representation , 2011, NIPS.

[15]  A. Ashok.  Stochastic Gradient Descent for Deep Learning , 2017 .

[16]  Alexandre Bernardino,et al.  Unifying Nuclear Norm and Bilinear Factorization Approaches for Low-Rank Matrix Decomposition , 2013, 2013 IEEE International Conference on Computer Vision.

[17]  Inderjit S. Dhillon,et al.  Scalable Coordinate Descent Approaches to Parallel Matrix Factorization for Recommender Systems , 2012, 2012 IEEE 12th International Conference on Data Mining.

[18]  Peter J. Haas,et al.  Large-scale matrix factorization with distributed stochastic gradient descent , 2011, KDD.

[19]  S. Du,et al.  Maxios : Large Scale Nonnegative Matrix Factorization for Collaborative Filtering , 2014 .

[20]  Pradeep Ravikumar,et al.  Collaborative Filtering with Graph Information: Consistency and Scalable Methods , 2015, NIPS.

[21]  Peder A. Olsen,et al.  Nuclear Norm Minimization via Active Subspace Selection , 2014, ICML.

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

[23]  Yehuda Koren,et al.  Matrix Factorization Techniques for Recommender Systems , 2009, Computer.