论文信息 - Analysis of a randomized approximation scheme for matrix multiplication

Analysis of a randomized approximation scheme for matrix multiplication

This note gives a simple analysis of a randomized approximation scheme for matrix multiplication proposed by Sarlos (2006) based on a random rotation followed by uniform column sampling. The result follows from a matrix version of Bernstein's inequality and a tail inequality for quadratic forms in subgaussian random vectors.

Sham M. Kakade | Tong Zhang | Daniel J. Hsu

[1] Sham M. Kakade,et al. An Analysis of Random Design Linear Regression , 2011, ArXiv.

[2] Sham M. Kakade,et al. A tail inequality for quadratic forms of subgaussian random vectors , 2011, ArXiv.

[3] Daniel J. Hsu,et al. Tail inequalities for sums of random matrices that depend on the intrinsic dimension , 2012 .

[4] Avner Magen,et al. Low rank matrix-valued chernoff bounds and approximate matrix multiplication , 2010, SODA '11.

[5] Bernard Chazelle,et al. Approximate nearest neighbors and the fast Johnson-Lindenstrauss transform , 2006, STOC '06.

[6] Nir Ailon,et al. Fast Dimension Reduction Using Rademacher Series on Dual BCH Codes , 2008, SODA '08.

[7] Christos Boutsidis,et al. Efficient Dimensionality Reduction for Canonical Correlation Analysis , 2012, SIAM J. Sci. Comput..

[8] Tamás Sarlós,et al. Improved Approximation Algorithms for Large Matrices via Random Projections , 2006, 2006 47th Annual IEEE Symposium on Foundations of Computer Science (FOCS'06).