Spectral properties of threshold functions


We examine the spectra of boolean functions obtained as the sign of a real polynomial of degreed. A tight lower bound on various norms of the lowerd levels of the function's Fourier transform is established. The result is applied to derive best possible lower bounds on the influences of variables on linear threshold functions. Some conjectures are posed concerning upper and lower bounds on influences of variables in higher order threshold functions.

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


  1. [1]

    D. H. Ackley, G. E. Hinton, andT. J. Sejnowski: A learning algorithm for Boltzmann machines,Cognitive Science 9 (1985), 147–169.

    Google Scholar 

  2. [2]

    R. Ahlswede, andZ. Zhang: An identity in combinatorial extremal theory,Advances in Mathematics 80 (1990), 137–151.

    Google Scholar 

  3. [3]

    J. Aspnes, R. Beigel, M. Furst, andS. Rudich: The expressive power of voting polynomials, in:Proceedings of the 23rd Symposium on the Theory of Computing, 402–409, ACM, 1991.

  4. [4]

    D. A. Mix Barrington, H. Straubing, andD. Thérien: Non-uniform automata over groups,Information and Computation 89 (1990), 109–132.

    Google Scholar 

  5. [5]

    M. Ben-Or, andN. Linial: Collective coin flipping, in:Randomness and Computation, (Micali S., editor). Academic Press, 1989.

  6. [6]

    J. Bourgain: Walsh subspaces ofl p product spaces, in:Seminaire D'Analyse Fonctionnelle, 4.1–4.9. Ecole Polytechnique, Centre de Mathematiques, 1979–1980.

  7. [7]

    J. Bruck: Harmonic analysis of polynomial threshold functions,SIAM Journal of Discrete Maths. 3 (1990), 268–177.

    Google Scholar 

  8. [8]

    J. Bruck: Polynomial threshold functions,AC 0 functions and spectral norms,SIAM Journal of Computing 21 (1) (1992), 33–42.

    Google Scholar 

  9. [9]

    C. Gotsman: On boolean functions, polynomials and algebraic threshold functions, Technical Report TR-89-18, Dept. of Computer Science, Hebrew University.

  10. [10]

    U. Haagerup: The best constants in the Khintchine inequality,Studia Mathematica 70 (1982), 231–283.

    Google Scholar 

  11. [11]

    J. Håstad: Personal communication, 1990.

  12. [12]

    J. J. Hopfield: Neural networks and physical systems with emergent collective computational abilities,Proc. Natl. Acad. Sc. USA 79 (1982), 2554–2558.

    Google Scholar 

  13. [13]

    J. Kahn, G. Kalai, andN. Linial: The influence of variables on boolean functions, in:Proceedings of the 29th Symposium on the Foundations of Computer Science, 68–80, IEEE, 1988.

  14. [14]

    J. Kahn, andR. Meshulam: On modp transversals,Combinatorica 11 (1) (1991), 17–22.

    Google Scholar 

  15. [15]

    Y. Katznelson:An Introduction to Harmonic Analysis, Wiley, 1968.

  16. [16]

    A. Kushilevitz, andY. Mansour: Learning decision trees using the Fourier spectrum, in:Proceedings of the 23rd Symposium on the Theory of Computing, 455–464, ACM, 1991.

  17. [17]

    Y. Le Cun: Une procedure d'apretissage pour reseau a seuil asymmetrique (A learning procedure for asymmetric threshold networks), in:Proceedings of Cognitiva 85, 599–604. Paris, 1985.

  18. [18]

    R. J. Lechner: Harmonic analysis of switching functions, in:Recent Developments in Switching Theory, (A. Mukhopadhyay, editor) Academic Press, 1971.

  19. [19]

    N. Linial, Y. Mansour, andN. Nisan: Constant depth circuits, Fourier transforms and learnability, in:Proceedings of the 30th Symposium on the Foundations of Computer Science, 574–579, IEEE, 1989.

  20. [20]

    M. Minsky, andS. Papert:Perceptrons: An Introduction to Computational Geometry, MIT Press, 1968.

  21. [21]

    I. Ninomiya: A theory of the coordinate representation of switching functions (review),IEEE Transactions on Electronic Computers 12 (1963), 152.

    Google Scholar 

  22. [22]

    P. O'Neil: Hyperplane cuts of ann-cube,Discrete Maths. 1 (1971), 193–195.

    Google Scholar 

  23. [23]

    M. Paterson: Personal communication, 1990.

  24. [24]

    R. Paturi, andM. Saks: On threshold circuits for parity, in:Proceedings of the 31st Symposium on the Foundations of Computer Science, 397–404, IEEE, 1990.

  25. [25]

    D. Rumelhart, G. Hinton, andJ. Williams: Learning internal representations by error propogation, in:Parallel Distributed Processing: Explorations in the Microstructure of Cognition, Vol 1., (Rumelhart J., McLelland J., and the PDP research group, editors) M.I.T. Press, 1986.

  26. [26]

    J. V. Uspensky:Introduction to Mathematical Probability, McGraw-Hill, 1937.

  27. [27]

    G. N. Watson: Theorems stated by Ramanujan (v): Approximations connected withe x,Proc. London Math. Society (2nd Series) 29 (1929), 293–308.

    Google Scholar 

  28. [28]

    R. Widner: Chow parameters in threshold logic,Journal of the ACM 18 (1971), 265–289.

    Google Scholar 

Download references

Author information



Additional information

Supported by an Eshkol fellowship, administered by the National Council for Research and Development—Israel Ministry of Science and Development.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Gotsman, C., Linial, N. Spectral properties of threshold functions. Combinatorica 14, 35–50 (1994). https://doi.org/10.1007/BF01305949

Download citation

AMS subject classification codes (1991)

  • 68 Q 15
  • 68 R 05