The words contained in this file might help you see if this file matches what you are looking for:

...Compressedlinearalgebraforlarge scale machinelearning ahmedelgohary matthias boehm peter j haas frederick r reiss berthold reinwald ibm research almaden san jose ca usa university of maryland college park md gb s per node uncompressed abstract data fits in time large machine learning ml algorithms are often execution memory operation performance compressed iterative using repeated read only access and i o space bound matrix vector multiplications to converge an opti compression ratio mal model it is crucial for t the into single or distributed main general purpose heavy lightweight techniques struggle size achieve both good ratios fast decompres figure goals linear algebra sion speed enable block wise operations scientists exibility create customize hence weinitiate work on cla independent cluster characteristics which database without worrying about underlying representations applied matrices then such e g sparse dense format plan generation as multiplication executed directly problem...