Stable distributions, pseudorandom generators, embeddings and data stream computation

In this paper we show several results obtained by combining the use of stable distributions with pseudorandom generators for bounded space. In particular: we show how to maintain (using only O(log n//spl epsiv//sup 2/) words of storage) a sketch C(p) of a point p/spl isin/l/sub 1//sup n/ under dynamic updates of its coordinates, such that given sketches C(p) and C(q) one can estimate |p-q|/sub 1/ up to a factor of (1+/spl epsiv/) with large probability. We obtain another sketch function C' which maps l/sub 1//sup n/ into a normed space l/sub 1//sup m/ (as opposed to C), such that m=m(n) is much smaller than n; to our knowledge this is the first dimensionality reduction lemma for l/sub 1/ norm we give an explicit embedding of l/sub 2//sup n/ into l/sub l//sup nO(log n)/ with distortion (1+1/n/sup /spl theta/(1)/) and a non-constructive embedding of l/sub 2//sup n/ into l/sub 1//sup O(n)/ with distortion (1+/spl epsiv/) such that the embedding can be represented using only O(n log/sup 2/ n) bits (as opposed to at least n/sup 2/ used by earlier methods).