Abstract
This paper investigates a link between matrix factorisation algorithms and recursive linear filters. In particular, we describe a probabilistic model in which sequential inference naturally leads to a matrix factorisation procedure. Using this probabilistic model, we derive a matrix-variate recursive linear filter that can be run efficiently in high-dimensional settings and leads to the factorisation of the data matrix into a dictionary matrix and a coefficient matrix. The resulting algorithm, referred to as the dictionary filter, is inherently online and has easy-to-tune parameters. We provide an extension of the proposed method for the cases where the dataset of interest is time-varying and nonstationary, thereby showing the adaptability of the proposed framework to non-standard problem settings. Numerical results, which are provided for image restoration and video modelling problems, demonstrate that the proposed method is a viable alternative to existing methods.
Similar content being viewed by others
Notes
The relative performance of the three methods remains similar when we change the percentage of missing data.
References
Lee, D.D., Seung, H.S.: Learning the parts of objects by non-negative matrix factorization. Nature 401(6755), 788–791 (1999)
Lee, D.D., Seung, H.S.: Algorithms for non-negative matrix factorization. In: NIPS, pp. 556–562, (2001)
Lin, C.-J.: Projected gradient methods for nonnegative matrix factorization. Neural Comput. 19(10), 2756–2779 (2007)
Koren, Y., Bell, R., Volinsky, C.: Matrix factorization techniques for recommender systems. Computer 1(8), 30–37 (2009)
Salakhutdinov, R., Mnih, A.: Probabilistic matrix factorization. In: Neural Information Processing Systems (NIPS) Conference, vol. 1, pp. 2–1, (2007)
Schmidt, M.N., Winther, O., Hansen, L.K.: Bayesian non-negative matrix factorization. In: International Conference on Independent Component Analysis and Signal Separation, pp. 540–547. Springer, (2009)
Cemgil, A.T.: Bayesian inference for nonnegative matrix factorisation models. Comput. Intell. Neurosci. 2009, 17 (2009). https://doi.org/10.1155/2009/785152
Bucak, S.S., Gunsel, B.: Incremental subspace learning via non-negative matrix factorization. Pattern Recognit. 42(5), 788–797 (2009)
Gemulla, R., Nijkamp, E., Haas, P.J., Sismanis, Y.: Large-scale matrix factorization with distributed stochastic gradient descent. In: Proceedings of the 17th ACM SIGKDD international conference on Knowledge discovery and data mining, pp. 69–77. ACM, (2011)
Mairal, J., Bach, F., Ponce, J., Sapiro, G.: Online learning for matrix factorization and sparse coding. J. Mach. Learn. Res. 11, 19–60 (2010)
Guan, N., Tao, D., Luo, Z., Yuan, B.: Online nonnegative matrix factorization with robust stochastic approximation. IEEE Trans. Neural Netw. Learn. Syst. 23(7), 1087–1099 (2012)
Duchi, J., Hazan, E., Singer, Y.: Adaptive subgradient methods for online learning and stochastic optimization. J. Mach. Learn. Res. 12, 2121–2159 (2011)
Kingma, D.P., Ba, J.: Adam: a method for stochastic optimization. In: International Conference for Learning Representations, (2015)
Mahsereci, M., Hennig, P.: Probabilistic line searches for stochastic optimization. In: Advances In Neural Information Processing Systems, pp. 181–189, (2015)
Yildirim, S., Cemgil, A.T., Singh, S.S.: An online expectation-maximisation algorithm for nonnegative matrix factorisation models. In: 16th IFAC Symposium on System Identification (SYSID 2012), (2012)
Paisley, J., Blei, D., Jordan, M.I.: Bayesian nonnegative matrix factorization with stochastic variational inference. In: volume Handbook of Mixed Membership Models and Their Applications, chapter 11. Chapman and Hall/CRC, (2015)
Carvalho, C.M., West, M., et al.: Dynamic matrix-variate graphical models. Bayesian Anal. 2(1), 69–97 (2007)
Triantafyllopoulos, K.: Reference priors for matrix-variate dynamic linear models. Commun. Stat.-Theory Methods 37(6), 947–958 (2008)
Hennig, P., Kiefel, M.: Quasi-newton methods: a new direction. J. Mach. Learn. Res. 14(1), 843–865 (2013)
Bertsekas, D.P.: Nonlinear programming. Athena scientific, (1999)
Ahn, S., Korattikara, A., Liu, N., Rajan, S., Welling, M.: Large-scale distributed Bayesian matrix factorization using stochastic gradient MCMC. In: Proceedings of the 21th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, pp. 9–18. ACM, (2015)
Harville, D.A.: Matrix algebra from a statistician’s perspective, vol. 1. Springer, (1997)
Anderson, B.D.O., Moore, J.B.: Optimal filtering. Englewood Cliffs 21, 22–95 (1979)
Bottou, L.: Online learning and stochastic approximations. In: Saad, D. (ed.) Online Learning and Neural Networks. Cambridge University Press, Cambridge, UK (1998)
Ollivier, Y.: Online natural gradient as a Kalman filter. arXiv:1703.00209, (2017). Accessed 30 Aug 2017
Bertsekas, D.P.: Incremental gradient, subgradient, and proximal methods for convex optimization: a survey. Optim. Mach. Learn. 2010, 1–38 (2011)
Akyıldız, Ö.D.: Online matrix factorization via Broyden updates. arXiv:1506.04389, (2015). Accessed 25 Aug 2017
Akyıldız, Ö.D., Elvira, V., Míguez, J.: The incremental proximal method: a probabilistic perspective. In: Proc. of IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP), pp. 4279–4283, (2018)
Abu-El-Haija, S., Kothari, N., Lee, J., Natsev, P., Toderici, G., Varadarajan, B., Vijayanarasimhan, S.: Youtube-8m: a large-scale video classification benchmark. arXiv:1609.08675, (2016). Accessed 12 Aug 2018
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Authors acknowledge the support of the Office of Naval Research Global (award no. N62909- 15-1-2011) and Ministerio de Economía y Competitividad of Spain (project TEC2015-69868-C2-1-R ADVENTURE) and the regional government of Madrid (program CASICAM-CM S2013/ICE-2845).
Rights and permissions
About this article
Cite this article
Akyildiz, Ö.D., Míguez, J. Dictionary filtering: a probabilistic approach to online matrix factorisation. SIViP 13, 737–744 (2019). https://doi.org/10.1007/s11760-018-1403-9
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11760-018-1403-9