Existence of Simple Extractors

We show that a small subset of seeds of any strong extractor also gives a strong extractor with similar parameters when the number of output bits is a constant. Specifically, if Ext : {0, 1} × {0, 1} → {0, 1} is a strong (k, )-extractor, then for at least 99% of choices of Õ(n ·2/ ) seeds, Ext restricted to these seeds is a (k, 3 )-extractor. Note that the degree of this restricted extractor is essentially optimal for m = O(1). By combining this with the Leftover Hash Lemma, we deduce that there are strong extractors outputting a constant number of bits with essentially optimal degree where each seed is a linear function, or even a Toeplitz matrix, or a simply-implementable hash function. Although linear extractors were known, such as the one by Trevisan [Tre01], it didn’t have close to optimal degree (although it did output more bits), and it wasn’t known that most sets of linear functions give extractors. While a simple application of the basic probabilistic method shows the existence of ordinary strong extractors, this approach fails to show the existence of the restricted extractors we seek, or even linear extractors. We therefore adopt a more sophisticated approach, using chaining as used by Rudra and Wootters [RW14] and others, combined with the Beck-Fiala theorem from discrepancy theory. ∗Supported by NSF Grant CCF-1526952, NSF Grant CCF-1705028, and a Simons Investigator Award (#409864, David Zuckerman). The authors began this research while attending the Pseudorandomness program at the Simons Institute for Theoretical Computer Science in Berkeley, CA. ISSN 1433-8092 Electronic Colloquium on Computational Complexity, Revision 1 of Report No. 116 (2018)

[1]  Jaikumar Radhakrishnan,et al.  Bounds for Dispersers, Extractors, and Depth-Two Superconcentrators , 2000, SIAM J. Discret. Math..

[2]  Christopher Umans,et al.  Simple extractors for all min-entropies and a new pseudorandom generator , 2005, JACM.

[3]  Luca Trevisan,et al.  Extractors and pseudorandom generators , 2001, JACM.

[4]  Larry Carter,et al.  Universal classes of hash functions (Extended Abstract) , 1977, STOC '77.

[5]  Xin Li,et al.  Improved Two-Source Extractors, and Affine Extractors for Polylogarithmic Entropy , 2016, 2016 IEEE 57th Annual Symposium on Foundations of Computer Science (FOCS).

[6]  Martti Penttonen,et al.  A Reliable Randomized Algorithm for the Closest-Pair Problem , 1997, J. Algorithms.

[7]  Ronen Shaltiel,et al.  Recent Developments in Explicit Constructions of Extractors , 2002, Bull. EATCS.

[8]  Ran Raz,et al.  Error reduction for extractors , 1999, 40th Annual Symposium on Foundations of Computer Science (Cat. No.99CB37039).

[9]  Emanuele Viola,et al.  On Randomness Extraction in AC0 , 2015, Electron. Colloquium Comput. Complex..

[10]  Enkatesan G Uruswami Unbalanced expanders and randomness extractors from Parvaresh-Vardy codes , 2008 .

[11]  Amnon Ta-Shma,et al.  An efficient reduction from two-source to non-malleable extractors: achieving near-logarithmic min-entropy , 2017, STOC.

[12]  Umesh V. Vazirani,et al.  Efficiency considerations in using semi-random sources , 1987, STOC.

[13]  Noam Nisan,et al.  Randomness is Linear in Space , 1996, J. Comput. Syst. Sci..

[14]  Atri Rudra,et al.  Every list-decodable code for high noise has abundant near-optimal rate puncturings , 2013, STOC.

[15]  Mary Wootters,et al.  On the list decodability of random linear codes with large error rates , 2013, STOC '13.

[16]  Bernard Chazelle,et al.  The Discrepancy Method , 1998, ISAAC.

[17]  David Zuckerman,et al.  Electronic Colloquium on Computational Complexity, Report No. 100 (2005) Linear Degree Extractors and the Inapproximability of MAX CLIQUE and CHROMATIC NUMBER , 2005 .

[18]  Eshan Chattopadhyay,et al.  Extractors for sumset sources , 2016, STOC.

[19]  Siyao Guo,et al.  Sparse extractor families for all the entropy , 2012, ITCS '13.

[20]  Venkatesan Guruswami,et al.  Restricted Isometry of Fourier Matrices and List Decodability of Random Linear Codes , 2013, SIAM J. Comput..

[21]  Kuan Cheng,et al.  Randomness Extraction in AC0 and with Small Locality , 2016, Electron. Colloquium Comput. Complex..

[22]  Miklos Santha,et al.  Generating Quasi-random Sequences from Semi-random Sources , 1986, J. Comput. Syst. Sci..

[23]  Leonid A. Levin,et al.  Pseudo-random generation from one-way functions , 1989, STOC '89.

[24]  M. Rudelson,et al.  On sparse reconstruction from Fourier and Gaussian measurements , 2008 .

[25]  Mikkel Thorup High Speed Hashing for Integers and Strings , 2015, ArXiv.

[26]  Philipp Woelfel,et al.  Efficient Strongly Universal and Optimally Universal Hashing , 1999, MFCS.