This paper proves that the additive error drops exponentially by iterating the sampling in an adaptive manner, and gives a pass-efficient algorithm for computing low-rank approximation with reduced additive error.Expand

We show that a simple spectral algorithm for learning a mixture of k spherical Gaussians in Rn works remarkably well--it succeeds in identifying the Gaussians assuming essentially the minimum… Expand

Modern data-mining methods are utilized, specifically classification trees and clustering algorithms, along with claims data from over 800,000 insured individuals over three years, to provide rigorously validated predictions of health-care costs in the third year, based on medical and cost data from the first two years.Expand

We show that a simple spectral algorithm for learning a mixture of k spherical Gaussians in /spl Ropf//sup n/ works remarkably well - it succeeds in identifying the Gaussians assuming essentially the… Expand

A divide-and-merge methodology for clustering a set of objects that combines a top-down "divide" phase with a bottom-up "merge" phase is presented and a meta-search engine that uses this methodology to cluster results from web searches is presented.Expand

A divide-and-merge methodology for clustering a set of objects that combines a top-down “divide" phase with a bottom-up “merge” phase is presented and the implementation of a meta-search engine that uses this methodology to cluster results from web searches is described.Expand

Frieze et al. [17] proved that a small sample of rows of a given matrix A contains a low-rank approximation D that minimizes ||A - D||F to within small additive error, and the sampling can be done… Expand

The first result shows that with adaptive sampling in t rounds and O(k/ ) samples in each round, the additive error drops exponentially as ; the computation time is nearly linear in the number of nonzero entries.Expand

This paper states that broad matching has undesirable economic properties (such as the non-existence of equilibria) that can make it hard for an advertiser to determine how much to bid for a broad-matched keyword.Expand

This work forms a family of graph covering problems whose goals are to suggest a subset of ads with the maximum benefit by suggesting rewrites for a given query by obtaining constant-factor approximation algorithms for these covering problems.Expand