Computing Symmetric Functions with AND/OR Circuits and a Single MAJORITY Gate

Fagin it et al. characterized those symmetric Boolean functions which can be computed by small AND/OR circuits of constant depth and unbounded fan-in. Here we provide a similar characterization for d-perceptron s --- AND/OR circuits of constant depth and unbounded fan-in with a single MAJORITY gate at the output. We show that a symmetric function has small (quasipolynomial, or 2 log n ^O(1) size) d-perceptrons iff it has only poly-log many sign changes (i.e., it changes value log^O(1) n times as the number of positive inputs varies from zero to n). A consequence of the lower bound is that a recent construction of Beigel is optimal. He showed how to convert a constant-depth unbounded fan-in AND/OR circuit with poly-log many MAJORITY gates into an equivalent d-perceptron --- we show that more than poly-log MAJORITY gates cannot in general be converted to one.

[1]  Ronald Fagin,et al.  Bounded-Depth, Polynomial-Size Circuits for Symmetric Functions , 1985, Theoretical Computer Science.

[2]  Zhi-Li Zhang.  Complexity of Symmetric Functions in Perceptron-Like Models , 1992 .

[3]  Daniel A. Spielman,et al.  The perceptron strikes back , 1991, [1991] Proceedings of the Sixth Annual Structure in Complexity Theory Conference.

[4]  J. Van Leeuwen,et al.  Handbook of theoretical computer science - Part A: Algorithms and complexity; Part B: Formal models and semantics , 1990 .

[5]  Ravi B. Boppana,et al.  The Complexity of Finite Functions , 1991, Handbook of Theoretical Computer Science, Volume A: Algorithms and Complexity.

[6]  J. Håstad.  Computational limitations of small-depth circuits , 1987 .

[7]  Noam Nisan,et al.  Constant depth circuits, Fourier transform, and learnability , 1989, 30th Annual Symposium on Foundations of Computer Science.

[8]  Andrew Chi-Chih Yao,et al.  Separating the Polynomial-Time Hierarchy by Oracles (Preliminary Version) , 1985, FOCS.

[9]  Nathan Linial,et al.  The influence of variables on Boolean functions , 1988, [Proceedings 1988] 29th Annual Symposium on Foundations of Computer Science.

[10]  Frederic Green.  An Oracle Separating +P From PP ph . , 1990 .

[11]  David A. Mix Barrington.  Quasipolynomial size circuit classes , 1992, [1992] Proceedings of the Seventh Annual Structure in Complexity Theory Conference.

[12]  A. Razborov.  Lower bounds on the size of bounded depth circuits over a complete basis with logical addition , 1987 .

[13]  Richard Beigel,et al.  The polynomial method in circuit complexity , 1993, [1993] Proceedings of the Eigth Annual Structure in Complexity Theory Conference.

[14]  Howard Straubing,et al.  Complex Polynomials and Circuit Lower Bounds for Modular Counting , 1992, LATIN.

[15]  A. Yao.  Separating the polynomial-time hierarchy by oracles , 1985 .

[16]  Jehoshua Bruck,et al.  Harmonic Analysis of Polynomial Threshold Functions , 1990, SIAM J. Discret. Math..

[17]  Richard Beigel.  When do extra majority gates help? , 1992, STOC '92.

[18]  Juris Hartmanis,et al.  One-Way Functions, Robustness, and the Non-Isomorphism of NP-Complete Sets , 1987, Proceeding Structure in Complexity Theory.

[19]  Roman Smolensky,et al.  Algebraic methods in the theory of lower bounds for Boolean circuit complexity , 1987, STOC.

[20]  Frederic Green.  An Oracle Separating \oplus P from PP^PH , 1991, Inf. Process. Lett..

[21]  Miklós Ajtai,et al.  ∑11-Formulae on finite structures , 1983, Ann. Pure Appl. Log..

[22]  Frederic Green.  An oracle separating (+)P from PP/sup PH/ , 1990, Proceedings Fifth Annual Structure in Complexity Theory Conference.

[23]  James Aspnes,et al.  The expressive power of voting polynomials , 1991, STOC '91.

[24]  Manuel Blum,et al.  Generic oracles and oracle classes , 1987, 28th Annual Symposium on Foundations of Computer Science (sfcs 1987).

[25]  MansourYishay,et al.  Constant depth circuits, Fourier transform, and learnability , 1993 .

[26]  Michael Sipser,et al.  Parity, circuits, and the polynomial-time hierarchy , 1981, 22nd Annual Symposium on Foundations of Computer Science (sfcs 1981).