Abstract
We propose an algorithm for Bayesian low-rank tensor completion with automatic rank determination in the canonical polyadic format when additional subspace information (SI) is given. We numerically validate the regularization properties induced by SI and present the results about tensor recovery and rank determination. The results show that the number of samples required for successful completion is significantly reduced in the presence of SI.
REFERENCES
D. L. Donoho, ‘‘Compressed sensing,’’ IEEE Trans. Inf. Theory 52, 1289–1306 (2006).
E. J. Candès and B. Recht, ‘‘Exact matrix completion via convex optimization,’’ Found. Comput. Math. 9, 717 (2009).
E. J. Candès and T. Tao, ‘‘The power of convex relaxation: Near-optimal matrix completion,’’ IEEE Trans. Inf. Theory 56, 2053–2080 (2010).
E. J. Candès, X. Li, Y. Ma, and J. Wright, ‘‘Robust principal component analysis?,’’ J. ACM 58 (3), 1–37 (2011).
H. Ma, H. Yang, M. R. Lyu, and I. King, ‘‘SoRec: Social recommendation using probabilistic matrix factorization,’’ in Proceedings of the Conference on Information and Knowledge Management CIKM 2008 (2008), pp. 931–940.
H. Xie, C. Li, R. Y. D. Xu, and K. Mengersen, ‘‘Robust kernelized bayesian matrix factorization for video background/foreground separation,’’ in Machine Learning, Optimization, and Data Science, 5th International Conference, LOD 2019, Siena, Italy, September 10–13, 2019 (2019), pp. 484–495.
N. Natarajan and I. S. Dhillon, ‘‘Inductive matrix completion for predicting gene–disease associations,’’ Bioinformatics 30 (12), 60–68 (2014).
P. Zakeri, J. Simm, A. Arany, S. ElShal, and Y. Moreau, ‘‘Gene prioritization using Bayesian matrix factorization with genomic and phenotypic side information,’’ Bioinformatics 34 (13), 447–456 (2018).
B. Güvenç Paltun, H. Mamitsuka, and S. Kaski, ‘‘Improving drug response prediction by integrating multiple data sources: Matrix factorization, kernel and network-based approaches,’’ Brief. Bioinform. 22, 346–359 (2021).
P. Jain and I. S. Dhillon, ‘‘Provable inductive matrix completion,’’ arXiv: 1306.0626 (2013).
M. Xu, R. Jin, and Z. Zhou, ‘‘Speedup matrix completion with side information: Application to multi-label learning,’’ in Proceedings of the Conference on Advances in Neural Information Processing Systems NIPS 2013 (2013), pp. 2301–2309.
Y. Kim and S. Choi, ‘‘Scalable variational bayesian matrix factorization with side information,’’ in Proceedings of the AISTATS 2014 (2014), pp. 493–502.
K. Chiang, C. Hsieh, and I. S. Dhillon, ‘‘Matrix completion with noisy side information,’’ in Proceedings of the Conference on Advances in Neural Information Processing Systems NIPS 2015 (2015), pp. 3447–3455.
K. Chiang, C. Hsieh, and I. S. Dhillon, ‘‘Robust principal component analysis with side information,’’ in Proceedings of the International Conference on Machine Learning ICML 2016 (2016), pp. 2291–2299.
T. G. Kolda and B. W. Bader, ‘‘Tensor decompositions and applications,’’ SIAM Rev. 51, 455–500 (2009).
I. V. Oseledets, ‘‘Tensor-train decomposition,’’ SIAM J. Sci. Comput. 33, 2295–2317 (2011).
A. Cichocki, D. Mandic, L. de Lathauwer, G. Zhou, Q. Zhao, C. Caiafa, and H. A. Phan, ‘‘Tensor decompositions for signal processing applications: From two-way to multiway component analysis,’’ IEEE Signal Process. Mag. 32, 145–163 (2015).
E. E. Papalexakis, C. Faloustos, and N. D. Sidiropoulos, ‘‘Tensors for data mining and data fusion: Models, applications, and scalable algorithms,’’ ACM Trans. Intell. Syst. Technol. 8 (2), 16:1–16:44 (2016).
N. D. Sidiropoulos, L. de Lathauwer, X. Fu, K. Huang, E. E. Papalexakis, and C. Faloustos, ‘‘Tensor decomposition for signal processing and machine learning,’’ IEEE Trans. Signal Process. 65, 3551–3582 (2017).
M. Signoretto, L. de Lathauwer, and J. A. K. Suykens, ‘‘Nuclear norms for tensors and their use for convex multilinear estimation,’’ Linear Algebra Appl. 43 (2010, in press).
S. Gandy, B. Recht, and I. Yamada, ‘‘Tensor completion and low-\(n\)-rank tensor recovery via convex optimization,’’ Inverse Probl. 27, 025010 (2011).
J. A. Bengua, H. N. Phien, H. D. Tuan, and M. N. Do, ‘‘Efficient tensor completion for color image and video recovery: Low-rank tensor train,’’ IEEE Trans. Image Process. 26, 2466–2479 (2017).
H. Rauhut, R. Schneider, and Ž. Stojanac, ‘‘Tensor completion in hierarchical tensor representation,’’ in Compressed Sensing and Its Applications: Proceedings of the MATHEON Workshop 2013 (2015), pp. 419–450.
L. Grasedyck, M. Kluge, and S. Krämer, ‘‘Variants of alternating least squares tensor completion in the tensor train format,’’ SIAM J. Sci. Comput. 37, A2424–A2450 (2015).
L. Grasedyck and S. Krämer, ‘‘Stable ALS approximation in the TT-format for rank-adaptive tensor completion,’’ Numer. Math. 143, 855–904 (2019).
D. Kressner, M. Steinlechner, and B. Vandereycken, ‘‘Low-rank tensor completion by Riemannian optimization,’’ BIT Numer. Math. 54, 447–468 (2014).
M. Steinlechner, ‘‘Riemannian optimization for high-dimensional tensor completion,’’ SIAM J. Sci. Comput. 38, S461–S484 (2016).
S. Budzinskiy and N. Zamarashkin, ‘‘Tensor train completion: Local recovery guarantees via riemannian optimization,’’ arXiv: 2110.03975 (2021).
V. de Silva and L. Lim, ‘‘Tensor rank and the ill-posedness of the best low-rank approximation problem,’’ SIAM J. Matrix Anal. Appl. 30, 1084–1127 (2008).
G. Tomasi and R. Bro, ‘‘PARAFAC and missing values,’’ Chemom. Intell. Lab. Syst. 75, 163–180 (2005).
E. Acar, D. M. Dunlavy, T. G. Kolda, and M. Mørup, ‘‘Scalable tensor factorizations for incomplete data,’’ Chemom. Intell. Lab. Syst. 106, 41–56 (2011).
T. Yokota, Q. Zhao, and A. Cichochki, ‘‘Smooth PARAFAC decomposition for tensor completion,’’ IEEE Trans. Signal Process. 64, 5423–5436 (2016).
R. Salakhutdinov and A. Mnih, ‘‘Probabilistic matrix factorization,’’ in Proceedings of the Conference on Advances in Neural Information Processing Systems NIPS 2007 (2008), Vol. 20.
R. M. Neal, Tech. Report CRG-TR-93-1 (Dep. Computer Sci., Univ. of Toronto, 1993).
M. J. Beal, Ph. D. Thesis (Univ. College, London, 2003).
J. Winn and C. M. Bishop, ‘‘Variational message passing,’’ J. Mach. Learn. Res. 6, 661–694 (2005).
R. Salakhutdinov and A. Mnih, ‘‘Bayesian probabilistic matrix factorization using Markov chain Monte Carlo,’’ in Proceedings of the International Conference on Machine Learning, ICML 2008 (2008), pp. 880–887.
D. E. Gilbert and M. T. Wells, ‘‘Tuning free rank-sparse bayesian matrix and tensor completion with global-local priors,’’ arXiv: 1905.11496 (2019).
B. Lakshminarayanan, G. Bouchard, and C. Archambeau, ‘‘Robust bayesian matrix factorisation,’’ in Proceedings of the International Conference on Artificial Intelligence and Statistics AISTATS 2011 (2011), pp. 425–433.
S. D. Babacan, M. Luessi, R. Molina, and A. K. Katsaggelos, ‘‘Sparse bayesian methods for low-rank matrix estimation,’’ IEEE Trans. Signal Process. 60, 3964–3977 (2012).
Y. Linxiao, J. Fang, H. Duan, H. Li, and B. Zeng, ‘‘Fast low-rank bayesian matrix completion with hierarchical gaussian prior models,’’ IEEE Trans. Signal Process. 66, 2804–2817 (2018).
W. Chu and Z. Ghahramani, ‘‘Probabilistic models for incomplete multi-dimensional arrays,’’ in Proceedings of the International Conference on Artificial Intelligence and Statistics AISTATS 2009 (2009), pp. 89–96.
Q. Zhao, L. Zhang, and A. Cichocki, ‘‘Bayesian sparse Tucker models for dimension reduction and tensor completion,’’ arXiv: 1505.02343 (2015).
L. Xu, L. Cheng, N. Wong, and Y. Wu, ‘‘Learning tensor train representation with automatic rank determination from incomplete noisy data,’’ arXiv: 2010.06564 (2020).
P. Rai, Y. Wang, S. Guo, G. Chen, D. Dunson, and L. Carin, ‘‘Scalable bayesian low-rank decomposition of incomplete multiway tensors,’’ in Proceedings of the International Conference on Machine Learning, ICML 2014 (2014), pp. 1800–1808.
Q. Zhao, L. Zhang, and A. Cichocki, ‘‘Bayesian CP factorization of incomplete tensors with automatic rank determination,’’ IEEE Trans. Pattern Anal. Mach. Intell. 37, 1751–1763 (2015).
Q. Zhao, G. Zhou, L. Zhang, A. Cichocki, and S. Amari, ‘‘Bayesian robust tensor factorization for incomplete multiway data,’’ IEEE Trans. Neural Network Learn. Syst. 27, 736–748 (2016).
L. Cheng, Z. Chen, Q. Shi, Y. Wu, and S. Theodoridis, ‘‘Towards probabilistic tensor canonical polyadic decomposition 2.0: Automatic tensor rank learning using generalized hyperbolic prior,’’ arXiv: 2009.02472 (2020).
P. Alquier, V. Cottet, N. Chopin, and J. Rousseau, ‘‘Bayesian matrix completion: Prior specification,’’ arXiv: 1406.1440 (2014).
E. Acar, T. G. Kolda, and D. M. Dunlavy, ‘‘All-at-once optimization for coupled matrix and tensor factorizations,’’ arXiv: 1105.3422 (2011).
A. Narita, K. Hayashi, R. Tomioka, and H. Kashima, ‘‘Tensor factorization using auxiliary information,’’ Data Mining Knowledge Discov. 25, 298–324 (2012).
T. Yokota, A. Cichocki, and Y. Yamashita, ‘‘Linked PARAFAC/CP tensor decomposition and its fast implementation for multi-block tensor analysis,’’ in Neural Information Processing, Proceedings of the Annual Conference (2012), pp. 84–91.
J. A. Bazerque, G. Mateos, and G. B. Giannakis, ‘‘Rank regularization and bayesian inference for tensor completion and extrapolation,’’ IEEE Trans. Signal Process. 61, 5689–5703 (2013).
Y. Wu, H. Tan, Y. Li, J. Zhang, and C. Xiaoxuan, ‘‘A fused CP factorization method for incomplete tensors,’’ IEEE Trans. Neural Network Learn. Syst. 30, 751–764 (2019).
Y. Guan, S. Dong, P. A. Absil, and F. Glineur, ‘‘Alternating minimization algorithms for graph regularized tensor completion,’’ arXiv: 2008.12876 (2020).
V. N. Ioannidis, A. S. Zamzam, G. B. Giannakis, and N. D. Sidiropoulos, ‘‘Coupled graphs and tensor factorization for recommender systems and community detection,’’ IEEE Trans. Knowledge Data Eng. 33, 909–920 (2021).
C. Yang, N. Singh, C. Xiao, C. Qian, E. Solomonik, and J. Sun, ‘‘MTC: Multiresolution tensor completion from partial and coarse observations,’’ in Proceedings of the ACM SIGKDD Conference on Knowledge Discovery and Data Mining KDD 2021 (2021), pp. 1953–1963.
B. Ermiş, E. Acar, and A. T. Cemgil, ‘‘Link prediction in heterogeneous data via generalized coupled tensor factorization,’’ Data Mining Knowledge Discov. 29, 203–236 (2015).
S. Budzinskiy and N. Zamarashkin, ‘‘Note: Low-rank tensor train completion with side information based on riemannian optimization,’’ arXiv: 2006.12798 (2020).
Z. Long, C. Zhu, J. Liu, P. Comon, and Y. Liu, ‘‘Trainable subspaces for low rank tensor completion: Model and analysis,’’ IEEE Trans. Signal Process. 70, 2502–2517 (2022).
S. Budzinskiy and N. Zamarashkin, ‘‘Variational bayesian inference for CP tensor completion with side information,’’ arXiv: 2206.12486 (2022).
F. Schäfer, M. Katzfuss, and H. Owhadi, ‘‘Sparse Cholesky factorization by Kullback–Leibler minimization,’’ SIAM J. Sci. Comput. 43, A2019–A2046 (2021).
Funding
This work was supported by Russian Science Foundation (project no. 21-71-10072).
Author information
Authors and Affiliations
Corresponding authors
Additional information
(Submitted by E. E. Tyrtyshnikov)
Rights and permissions
About this article
Cite this article
Budzinskiy, S., Zamarashkin, N. Variational Bayesian Inference for CP Tensor Completion with Subspace Information. Lobachevskii J Math 44, 3016–3027 (2023). https://doi.org/10.1134/S1995080223080103
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1134/S1995080223080103