Shrinkage of De Morgan Formulae from Quantum Query Complexity

We give a new and improved proof that the shrinkage exponent of De Morgan formulae is 2. Namely, we show that for any Boolean function f : {−1, 1} → {−1, 1}, setting each variable out of x1, . . . , xn with probability 1 − p to a randomly chosen constant, reduces the expected formula size of the function by a factor of O(p). This result is tight and improves the work of H̊astad [H̊as98] by removing logarithmic factors. As a consequence of our results, the function defined by Andreev [And87], A : {−1, 1} → {−1, 1}, which is in P, has formula size at least Ω( n 3 log2 n log3 logn ). This lower bound is tight up to the log log n factor, and is the best known lower bound for functions in P. In addition, the functions defined in [KRT13], hr : {−1, 1} → {−1, 1}, which are also in P, cannot be computed correctly on a fraction greater than 1/2 + 2−r of the inputs, by De Morgan formulae of size at most n 3 r2poly logn , for any parameter r ≤ n . The proof relies on a result from quantum query complexity by [LLS06, HLS07, Rei11]: for any Boolean function f , Q2(f) ≤ O( √ L(f)), where Q2(f) is the bounded-error quantum query complexity of f , and L(f) is the minimal size De Morgan formula computing f . ∗Department of Computer Science and Applied Mathematics, Weizmann Institute of Science, Rehovot 76100, Israel. avishay.tal@weizmann.ac.il. Supported by an Adams Fellowship of the Israel Academy of Sciences and Humanities, by an ISF grant and by the I-CORE Program of the Planning and Budgeting Committee. ISSN 1433-8092 Electronic Colloquium on Computational Complexity, Report No. 48 (2014)

[1]  Troy Lee,et al.  Negative weights make adversaries stronger , 2007, STOC '07.

[2]  Valentine Kabanets,et al.  An Improved Deterministic #SAT Algorithm for Small de Morgan Formulas , 2014, Algorithmica.

[3]  V. M. Khrapchenko.  Method of determining lower bounds for the complexity of P-schemes , 1971 .

[4]  Ben Reichardt,et al.  Span-program-based quantum algorithm for evaluating formulas , 2007, Theory Comput..

[5]  Andris Ambainis,et al.  Any AND-OR Formula of Size N can be Evaluated in time N^{1/2 + o(1)} on a Quantum Computer , 2010, 48th Annual IEEE Symposium on Foundations of Computer Science (FOCS'07).

[6]  Ronald de Wolf,et al.  Quantum lower bounds by polynomials , 1998, Proceedings 39th Annual Symposium on Foundations of Computer Science (Cat. No.98CB36280).

[7]  Troy Lee,et al.  The Quantum Adversary Method and Classical Formula Size Lower Bounds , 2005, Computational Complexity Conference.

[8]  David Zuckerman,et al.  Mining Circuit Lower Bound Proofs for Meta-algorithms , 2014, Computational Complexity Conference.

[9]  Ben Reichardt,et al.  Reflections for quantum query algorithms , 2010, SODA '11.

[10]  Noam Nisan,et al.  Constant depth circuits, Fourier transform, and learnability , 1993, JACM.

[11]  Ran Raz,et al.  Average-case lower bounds for formula size , 2013, STOC '13.

[12]  T. Sanders,et al.  Analysis of Boolean Functions , 2012, ArXiv.

[13]  Ran Raz,et al.  Improved Average-Case Lower Bounds for DeMorgan Formula Size , 2013, 2013 IEEE 54th Annual Symposium on Foundations of Computer Science.

[14]  Andrew M. Childs,et al.  ANY AND-OR FORMULA OF SIZE N CAN BE EVALUATED IN TIME N1/2+o(1) ON A QUANTUM COMPUTER , 2010 .

[15]  Ryan O'Donnell,et al.  Analysis of Boolean Functions , 2014, ArXiv.

[16]  Troy Lee.  A New Rank Technique for Formula Size Lower Bounds , 2007, STACS.

[17]  Rahul Santhanam,et al.  Fighting Perebor: New and Improved Algorithms for Formula and QBF Satisfiability , 2010, 2010 IEEE 51st Annual Symposium on Foundations of Computer Science.

[18]  Russell Impagliazzo,et al.  Pseudorandomness from Shrinkage , 2012, 2012 IEEE 53rd Annual Symposium on Foundations of Computer Science.

[19]  Ben Reichardt,et al.  Span Programs and Quantum Query Complexity: The General Adversary Bound Is Nearly Tight for Every Boolean Function , 2009, 2009 50th Annual IEEE Symposium on Foundations of Computer Science.

[20]  Noam Nisan,et al.  The Effect of Random Restrictions on Formula Size , 1993, Random Struct. Algorithms.

[21]  Ran Raz,et al.  Super-logarithmic depth lower bounds via the direct sum in communication complexity , 1995, computational complexity.

[22]  Or Meir,et al.  Toward better formula lower bounds: an information complexity approach to the KRW composition conjecture , 2014, STOC.

[23]  Russell Impagliazzo,et al.  Fourier Concentration from Shrinkage , 2014, Computational Complexity Conference.

[24]  Ronald de Wolf,et al.  Robust Polynomials and Quantum Algorithms , 2003, Theory of Computing Systems.

[25]  Johan Hå stad.  The Shrinkage Exponent of de Morgan Formulas is 2 , 1998 .

[26]  Uri Zwick,et al.  Shrinkage of de Morgan formulae under restriction , 1991, [1991] Proceedings 32nd Annual Symposium of Foundations of Computer Science.

[27]  Edward Farhi,et al.  A Quantum Algorithm for the Hamiltonian NAND Tree , 2008, Theory Comput..

[28]  R. Kaas,et al.  Mean, Median and Mode in Binomial Distributions , 1980 .