Evaluating spectral norms for constant depth circuits with symmetric gates

Abstract

The Fourier spectrum and its norms are given as explicit arithmetic expressions and evaluated, for Boolean functions computed by classes of constant depth, read-once circuits consisting of an arbitrary set of symmetric gates. Previous results of this nature estimate the spectralL 1 norm of functions computed by certain types of decision trees [20], [7], and in some cases, give randomizedprocedures that evaluate the spectrum by clever rounding [20]. One corollary of our results provides a large class ofAC 0 functions whose spectralL 1 norm is exponential, thus generalizing the single example of such a function given in [9]. This shows that almost every read-onceAC 0 function does not belong in the classPL 1 of functions with polynomially bounded spectral norms.

Implications of our results and technique are discussed, for estimating the spectral norms ofany function in a constant depth circuit class, using the coding theoretic concept of weight distributions. Evaluating the spectral norms for any such function reduces to estimating certain non-trivial weight distributions of simple, linear codes.

This is a preview of subscription content, access via your institution.

References

  1. [1]

    E. Allender, A note on the power of threshold circuits. InProc. 30th Ann. IEEE Symp. Foundations of Computer Science, 1989, 580–584.

  2. [2]

    D. A. Barrington, Bounded width, polynomial size branching programs recognize exactly those languages inNC. InProc. 18th Ann. ACM Symp. Theory of Computing, 1986, 1–5.

  3. [3]

    R. Beigel, “Why do extra majority gates help?” InProc. 24th Ann. ACM Symp. Theory of Computing, 1992, 450–454.

  4. [4]

    D. A. Barrington, R. Beigel, and S. Rudich, Representing Boolean functions as polynomials modulo composite numbers. InProc. 24th Ann. ACM Symp. Theory of Computing, 1992, 455–461.

  5. [5]

    Y. Brandman, A. Orlitsky, and J. Hennessy, A spectral lower bound technique for the size of decision trees and two-level and-or circuits. InIEEE Transaction on Computers 39:2 (1990), 282–287.

  6. [6]

    J. Bruck, Harmonic analysis of polynomial threshold functions. InSIAM Journal of Discrete Mathematics 3:2 (1990), 168–177.

  7. [7]

    M. Bellare, A technique for upper bounding the spectral norm with applications to learning. InProc. 5th Ann. IEEE Symp. Computational Learning Theory, 1992, 62–70.

  8. [8]

    R. Boppana and M. Sipser,The complexity of finite functions. Technical Report MIT/LCS/TM-405, Massachussetts Institute of Technology, Laboratory for Computer Sciences, 1989.

  9. [9]

    J. Bruck andR. Smolensky, Polynomial Threshold Functions,AC 0 Functions and Spectral Norms.SIAM Journal of Computing 21:1 (1992), 33–42.

    Google Scholar 

  10. [10]

    H. Dym and H. P. McKean,Fourier series and integrals. Probability and Methematical Statistics series, Academic Press, 1972.

  11. [11]

    M. Dowd and M. Sitharam, Shannon bounds for functions over ℤ 2 n . Technical Report 90-12-1, Kent State University, Department of Mathematics and Computer Science, 1990.

  12. [12]

    M. Furst, J. Jackson, and S. Smith, Improved learning ofAC 0 functions. InProc. 5th Ann. IEEE Symp. Computational Learning Theory, 1992, 317–325.

  13. [13]

    M. Furst, J. Saxe, andM. Sipser, Parity, circuits and the polynomial time hierarchy.Mathematical Systems Theory 17 (1984), 17–27.

    Google Scholar 

  14. [14]

    M. Goldman, J. Hastad, andA. A. Razborov, Majority gates vs. general weighted thresgold gates.Computational Complexity 2 (1992), 277–300.

    Google Scholar 

  15. [15]

    M. Goldman and J. Hastad, On the power of small depth threshold circuits. InProc. 31st Ann. IEEE Symp. Foundations of Computer Science, 1990, 610–618.

  16. [16]

    V. Grolmusz,Harmonic analysis, real approximation and communication complexity of Boolean functions. Manuscript, 1994.

  17. [17]

    J. Hastad,Computational limitations of small depth circuits. Ph. D thesis, Massachussetts Institute of Technology press, 1986.

  18. [18]

    A. Hajnal, W. Maass, P. Pudlák, M. Szegedy, and G. Turán, Threshold circuits of bounded depth. InProc. 28th Ann. IEEE Symp. Foundations of Computer Science, 1987, 99–110.

  19. [19]

    J. Kahn, J. Kalai, and N. Linial, The influence of variables on Boolean functions. InProc. 29th Ann. IEEE Symp. Foundations of Computer Science, 1988, 68–80.

  20. [20]

    E. Kushilevitz andY. Mansour, Learning decision trees using the Fourier transform.SIAM Journal of Computing 22:6 (1993), 1331–1348.

    Google Scholar 

  21. [21]

    M. Krause, Geometric arguments yield better bounds for threshold circuits and distributed computing. InProc. 6th Ann. IEEE Symp. Structure in Complexity Theory, 1991, 314–322.

  22. [22]

    M. Krause and S. Waack, Variation ranks of communication matrices and lower bounds for depth two circuits having symmetric gates and unbounded fanin. InProc. 32nd Ann. IEEE Symp. Foundations of Computer Science, 1991, 777–782.

  23. [23]

    N. Linial, Y. Mansour, and N. Nisan, Constant depth circuits, Fourier transforms, and learnability. InProc. 30th Ann. IEEE Symp. Foundations of Computer Science, 1989, 574–579. To appear inJACM.

  24. [24]

    N. Linial andN. Nisan, Approximate inclusion-exclusion.Combinatorica 10:4 (1990), 349–365.

    Google Scholar 

  25. [25]

    F. J. MacWiliams and N. J. A. Sloane,The theory of error-correcting codes. North Holland, 1977.

  26. [26]

    N. Nisan and A. W. Widgerson, Hardness vs. randomness. InProc. 29th Ann. IEEE Symp. Foundations of Computer Science, 1988, 2–11. To appear inJCSS.

  27. [27]

    N. Nisan and M. Szegedy, On the degree of Boolean functions as real polynomials. InProc. 24th Ann. ACM Symp. Theory of Computing, 1992, 468–467.

  28. [28]

    R. Paturi, On the degree of polynomials that approximate symmetric Boolean functions. InProc. 24th Ann. ACM Symp. Theory of Computing 1992, 468–474.

  29. [29]

    R. Paturi and M. Saks, Threshold circuits for parity. InProc. 31st Ann. IEEE Symp. Foundations of Computer Science, 1990, 397–404.

  30. [30]

    A. A. Razborov, Lower bounds on the monotone complexity of some Boolean functions.Soviet Mathematics Doklady 31 (1985), 354–357.

    Google Scholar 

  31. [3]

    A. A. Razborov, Lower bounds on the size of circuits of bounded depth with basis {∧,⊕}. InMath. Notes of the Academy of Sciences of the USSR 41:4 (1987), 333–338.

  32. [32]

    M. Sitharam, Pseudorandom generators and learning algorithms forAC 0. InProc. 26th Ann. ACM Symp. Theory of Computing, 1994, 478–488. To appear inComputational Complexity.

  33. [33]

    R. Smolensky, Algebraic methods in the theory of lower bounds for boolean circuit complexity. InProc. 19th Ann. ACM Symp. Theory of Computing, 1987, 77–82.

  34. [34]

    K. I. Siu andJ. Bruck, On the power of threshold circuits with small weights.SIAM Journal of Discrete Mathematics 4:3 (1991), 423–435.

    Google Scholar 

  35. [35]

    A. C. Yao, Lower bounds by probabilistic arguments. InProc. 24th Ann. IEEE Symp. Foundations of Computer Science, 1983, 420–428.

  36. [36]

    A. C. Yao, Separating the polynomial time hierarchy by oracles. InProc. 26th Ann. IEEE Symp. Foundations of Computer Science, 1985, 1–10.

  37. [37]

    A. C. Yao, Circuits and local computation. InProc. 21st Ann. ACM Symp. Theory of Computing, 1989, 186–196.

  38. [38]

    A. C. Yao, OnACC and threshold circuits. InProc. 31st Ann. IEEE Symp. Foundations of Computer Science, 1990, 619–627.

Download references

Author information

Affiliations

Authors

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Sitharam, M. Evaluating spectral norms for constant depth circuits with symmetric gates. Comput Complexity 5, 167–189 (1995). https://doi.org/10.1007/BF01268144

Download citation

Key words

  • Circuit complexity
  • lower bounds
  • Fourier transforms

Subject classifications

  • 68Q15
  • 68Q99