Original Paper

computational complexity

, Volume 14, Issue 3, pp 186-227

First online:

Open Access This content is freely available online to anyone, anywhere at any time.

Compression of Samplable Sources

  • Luca TrevisanAffiliated withComputer Science Division, U.C. Berkeley Email author 
  • , Salil VadhanAffiliated withDivision of Engineering and Applied Sciences, Harvard University
  • , David ZuckermanAffiliated withDepartment of Computer Science, University of Texas


We study the compression of polynomially samplable sources. In particular, we give efficient prefix-free compression and decompression algorithms for three classes of such sources (whose support is a subset of {0, 1} n ).
  1. 1.

    We show how to compress sources X samplable by logspace machines to expected length H(X) + O(1).

Our next results concern flat sources whose support is in P.
  1. 2.

    If H(X) ≤ k =  n − O(log n), we show how to compress to expected length k + polylog(nk).

  2. 3.

    If the support of X is the witness set for a self-reducible NP relation, then we show how to compress to expected length H(X) + 5.



Expander graphs arithmetic coding randomized logspace pseudorandom generators approximate counting

Subject classification.