Areas of computer science, such as randomized algorithms and cryptography,
rely on randomness. We present a survey paper by Nisan on a family of graphs
called dispersers and extractors. Because dispersers and extractors have
"random-like" characteristics, they can be used to extract an almost random
distribution from a "somewhat random" distribution. We provide an
introduction to this family of graphs, as well as motivation for studying
them.
Last modified 16 November 2002