Doing the Impossible (Almost): A Survey of Approximation Algorithms that Make Queries Vastly Faster
Computing various quantities such as medians or the number of unique elements requires a lot of time, a lot of memory, or both. It is, however, possible to get really close to the exact answer with much less time and much less memory. Some of these algorithms are much simpler than you might expect. Ted will describe a selection of these algorithms including some not yet published results. Ted will also outline how these algorithms can be applied to practical problems like anomaly detection.
Ted Dunning is Chief Application Architect at MapR Technologies and committer and PMC member for the Apache Mahout, Drill and Zookeeper projects and mentor for the Storm and Spark projects.
He was the chief architect behind the MusicMatch, (now Yahoo Music) and Veoh recommendation systems and built fraud detection systems for ID Analytics.