S Wang, J Liu, N Shroff - International Conference on …, 2018 - proceedings.mlr.press
… In this paper, we develop a new codedcomputation strategy, we call sparsecode, which achieves near optimal re… We also consider 8 sparsematrices from real data sets (Davis & Hu…
AB Das, A Ramamoorthy - IEEE Journal on Selected Areas in …, 2022 - ieeexplore.ieee.org
… fraction of matrix B, we propose a codedmatrix-matrix … partial computations done by the slower worker nodes; and (iii) enhances the worker computation speed when the “input” matrices …
… arrays used to store non-zero elements of sparsematrices or information about the non-zero … The first task that has to be solved by a developer of a sparsematrixcode is to determine …
J Willcock, A Lumsdaine - Proceedings of the 20th annual international …, 2006 - dl.acm.org
… are important for many scientific computations, with matrix-… sparsematrices, the primary performance limitation on matrix-… code for each matrix, but does not directly encode the matrix …
DJC MacKay, RM Neal - … Conference on Cryptography and Coding, 1995 - Springer
… The ideas behind MN codes are (1) that we use a sparse … matrix in terms of invertible matrices, such that the sparse … To compute the probability of error for BCH codes we evaluated the …
… Sparsematrices are widely used in scientific computation, especially in largescale optimization, structural and circuit analysis, computational fluid … adapted any of their code. MATLAB is …
… from the Harwell-Boeing collection and industrial application codes. Figure 1 summarizes … tured sparsematrices. These data structures are appropriate in cases where the sparsematrix …
… of a tool package for manipulating and working with sparsematrices. One of the goals of the … in sparsematrixcomputations. The starting point is the Harwell/Boeing collection of matrices …
FS Smailbegovic, GN Gaydadjiev… - Proceedings of the 16th …, 2005 - researchgate.net
… on SparseMatrices are the key computational kernels in many … of operations on sparse matrices we propose a simple and … -codedmatrix into blocks of SxS, and take for each SxS block …