This talk is an interim report on the 5 year plan launched in 1996 to provide a theoretical and computational foundation of Statistics for massive data sets. The plan coincided with the formation of AT&T Labs and the proposed research agenda of the InfoLab, which is both a physical laboratory and an interdisciplinary collection of information researchers in CS, mathematics, and Statistics. At the halfway point of this odyssey we can identify some success stories but more importantly it is an opportune time to re-calibrate the challenges and the milestones.
Unfortunately, ACM prohibits us from displaying non-influential references for this paper.
To see the full reference list, please visit http://dl.acm.org/citation.cfm?id=312183.