Abstract
Low-rank matrix approximation is a crucial technique for data analysis and scientific computing, and the Nyström method is one of the efficient sampling-based low-rank approximation schemes for handling large kernel matrices. The approximation accuracy of Nyström approach highly depends on the number of columns of the subset used, and it consumes much time on large data sets. This paper presents an accurate and fast Nyström approach to reducing the computational burdens when handling large kernel matrices, and experimental results show its competitive performance in both accuracy and efficiency.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsPreview
Unable to display preview. Download preview PDF.
References
Ng, A.Y., Jordan, M.I., Weiss, Y.: On spectral clustering: Analysis and an algorithm. In: Advances in Neural Information Processing Systems. MIT, Cambridge (2002)
Fowlkes, C., Belongie, S., Chung, F., Malik, J.: Spectral grouping using the nyströom method. IEEE Transactions on Pattern Analysis and Machine Intelligence 26(2), 214–225 (2004)
Vapnik, V.: The nature of statistical learning theory. Springer, New York (1995)
Belkin, M., Niyogi, P.: Laplacian eigenmaps and spectral techniques for embedding and clustering. In: Advances in Neural Information Processing Systems, pp. 585–591. MIT, Cambridge (2002)
Williams, C.K.I., Seeger, M.: Using the nyströom method to speed up kernel machines. In: Advances in Neural Information Processing Systems, pp. 682–688. MIT, Cambridge (2001)
Kumar, S., Mohri, M., Talwalkar, A.: Sampling techniques for the nyström method. In: The 12th International Conference on Artificial Intelligence and Statistics, pp. 304–311 (2009)
Kumar, S., Mohri, M., Talwalkar, A.: On sampling-based approximate spectral decomposition. In: International Conference on Machine Learning, pp. 553–560 (2009)
Zhang, K., Tsang, I., Kwok, J.: Improved nyström low-rank approximation and error analysis. In: International Conference on Machine Learning, pp. 1232–1239 (2008)
Kumar, S., Talwalkar, A.: Ensemble nyström method. In: Advances in Neural Information Processing Systems, pp. 1060–1068. MIT, Cambridge (2009)
Li, M., Kwok, J.T., Lu, B.-L.: Making large-scale nystrom approximation possible. In: International Conference on Machine Learning, pp. 1–8 (2010)
Rokhlin, V., Szlam, A., Tygert, M.: A randomized algorithm for principal component analysis. SIAM Journal on Matrix Analysis and Applications, 1100–1124 (2009)
Drineas, P., Mahoney, M.: On the nyström method for approximating a gram matrix for improved kernel-based learning. Journal of Machine Learning Research, 2153–2175 (2005)
Drineas, P., Mahoney, M.: Fast monte carlo algorithms for matrices II: Computing a low-rank approximation to a matrix. SIAM Journal on Computing 36(1), 158–183 (2006)
Lutkepohl, H.: Handbook of Matrices. John Wiley and Sons (1996)
Bhatia, R.: Matrix Analysis. Springer, New York (1997)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2012 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Zhang, H., Wang, Z., Cao, L. (2012). Fast Nyström for Low Rank Matrix Approximation. In: Zhou, S., Zhang, S., Karypis, G. (eds) Advanced Data Mining and Applications. ADMA 2012. Lecture Notes in Computer Science(), vol 7713. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-35527-1_38
Download citation
DOI: https://doi.org/10.1007/978-3-642-35527-1_38
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-35526-4
Online ISBN: 978-3-642-35527-1
eBook Packages: Computer ScienceComputer Science (R0)