Concentration of measure and isoperimetric inequalities in product spaces

  • Michel Talagrand
Article

Abstract

The concentration of measure phenomenon in product spaces roughly states that, if a set A in a product ΩN of probability spaces has measure at least one half, “most” of the points of Ωn are “close” to A. We proceed to a systematic exploration of this phenomenon. The meaning of the word “most” is made rigorous by isoperimetrictype inequalities that bound the measure of the exceptional sets. The meaning of the work “close” is defined in three main ways, each of them giving rise to related, but different inequalities. The inequalities are all proved through a common scheme of proof. Remarkably, this simple approach not only yields qualitatively optimal results, but, in many cases, captures near optimal numerical constants. A large number of applications are given, in particular to Percolation, Geometric Probability, Probability in Banach Spaces, to demonstrate in concrete situations the extremely wide range of application of the abstract tools.

AMS Classification numbers

Primary 60E15, 28A35, 60G99 Secondary 60G15, 68C15 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. [A-L-R]
    M. Aizenman, J. L. Lebowitz, D. Ruelle, Some rigorous results on the Sherrington-Kirkpatrick spin glass model,Commun. Math. Phys. 112 (1987), 3–20.MATHCrossRefMathSciNetGoogle Scholar
  2. [A-S]
    N. Alon, J. Spencer,The Probabilistic Method, Wiley, 1991.Google Scholar
  3. [A-M]
    D. Amir, V. D. Milman, Unconditional and symmetric sets inn-dimensional normed spaces,Israel J. Math. 37 (1980), 3–20.MATHMathSciNetGoogle Scholar
  4. [B1]
    B. Bollobás, The chromatic number of random graphs,Combinatorica 8 (1988), 49–55.MATHCrossRefMathSciNetGoogle Scholar
  5. [B2]
    B. Bollabás, Random graphs revisited,Proceedings of Symposia on Applied Mathematics, Vol. 44, 1991, 81–98.Google Scholar
  6. [B-B]
    B. Bollobás, G. Brightwell, The height of a random partial order: Concentration of Measure,Annals of Applied Probab.2 (1992), 1009–1018.MATHGoogle Scholar
  7. [C-L]
    E. G. Coffman, Jr.,G. S. Lucker,Probabilistic Analysis of Packing and Partitioning Algorithms, Wiley, 1991.Google Scholar
  8. [C-N]
    F. Comets, J. Neveu, The Sherrington-Kirkpatrick Model of Spin Classes and Stochastic Calculus: the high temperature case,Comm. Math. Phys. 166 (1995), 549–564.MATHCrossRefMathSciNetGoogle Scholar
  9. [D-MS]
    S. Dilworth, S. Montgomery-Smith, The distribution of vector-valued Rademacher series,Ann. Probab. 21 (1993), 2046–2052.MATHMathSciNetGoogle Scholar
  10. [F]
    A. M. Frieze, On the length of the longest monotone subsequence in a random permutation,Ann. Appl. Prob. 1 (1991), 301–305.MATHMathSciNetGoogle Scholar
  11. [G-M]
    M. Gromov, V. D. Milman, A topological application of the isoperimetric inequality,Amer. J. Math. 105 (1983), 843–854.MATHCrossRefMathSciNetGoogle Scholar
  12. [Har]
    L. H. Harper, Optimal numbering and isoperimetric problems on graphs,J. Comb. Theory (1966), 385–395.Google Scholar
  13. [H]
    W. Hoeffding, Probability inequalities for sums of bounded random variables,J. Amer. Statist. Assoc. 58 (1963), 13–30.MATHCrossRefMathSciNetGoogle Scholar
  14. [J]
    S. Janson, Poisson approximation for large deviations,Random Structures and Algorithms 1 (1990), 221–290.MATHCrossRefMathSciNetGoogle Scholar
  15. [J-S]
    W. Johnson, G. Schechtman, Remarks on Talagrand’s deviation inequality for Rademacher’s functions,Lecture Notes in Math. 1470, Springer Verlag, 1991, 72–77.MathSciNetGoogle Scholar
  16. [Ka]
    R. M. Karp, An upper bound on the expected cost of an optimal assignment, inDiscrete Algorithm and Complexity: Proceedings of the Japan-US joint Seminar, Academic Press, 1987, 1–4.Google Scholar
  17. [K1]
    H. Kesten, Aspects of first-passage percolation, Ecole d’Eté de Probabilité de Saint-Flour XIV,Lecture Notes in Math. 1180, 125–264, Springer Verlag, 1986, 125–264.MathSciNetGoogle Scholar
  18. [K2]
    H. Kesten, On the speed of convergence in first passage percolation,Ann. Applied Probab. 3 (1993), 296–338.MATHMathSciNetGoogle Scholar
  19. [K-S]
    R. M. Karp, J. M. Steele, Probabilistic analysis of heuristics, inThe Traveling Salesman Problem, John Wiley and Sons, 1985, 181–205.Google Scholar
  20. [Lea]
    J. Leader, Discrete isoperimetric inequalities,Proceedings of Symposia on Applied Mathematics, Vol. 44, 1991, 57–80.MathSciNetGoogle Scholar
  21. [L]
    M. Ledoux,Gaussian randomization and the law of the iterated logarithm in type 2 Banach spaces, Unpublished manuscript, 1985.Google Scholar
  22. [L-T1]
    M. Ledoux, M. Talagrand, Characterization of the law of the iterated logarithm in Banach spaces,Ann. Probab. 16 (1988), 1242–1264.MATHMathSciNetGoogle Scholar
  23. [L-T2]
    M. Ledoux, M. Talagrand,Probability in Banach Spaces, Springer Verlag, 1991.Google Scholar
  24. [Lu]
    T. Luczak, The chromatic number of Random graphs,Combinatorica 11 (1991), 45–54.MATHCrossRefMathSciNetGoogle Scholar
  25. [Mau1]
    B. Maurey, Construction de suites symétriques,Comptes Rendus Acad. Sci. Paris 288 (1979), 679–681.MATHMathSciNetGoogle Scholar
  26. [Mau2]
    B. Maurey, Some deviation inequalities,Geometric and Functional Analysis 1 (1991), 188–197.MATHCrossRefMathSciNetGoogle Scholar
  27. [McD]
    C. McDiarmid, On the method of bounded differences, inSurvey in Combinatorics (J. Simons, Ed.), London Mathematical Society Lecture Notes, Vol. 141, Cambridge Univ. Press, London/New York, 1989, 148–188.Google Scholar
  28. [M-H]
    C. McDiarmid, RyanHayward, Strong concentration for Quicksort,Proceedings of the Third Annual ACM-SIAM Symposium on Discrete Algorithms (SODA), 1992, 414–421.Google Scholar
  29. [M-S]
    V. D. Milman, G. Schechtman, Asymptotic theory of finite dimensional normed spaces,Lecture Notes in Math. 1200, Springer Verlag, 1986.Google Scholar
  30. [Mi1]
    V. D. Milman, A new proof of the theorem of A. Dvoretzky on sections of convex bodies,Func. Anal. Appl. 5 (1971), 28–37.MathSciNetGoogle Scholar
  31. [Mi2]
    V. D. Milman, Asymptotic properties of functions of several variables defined on homogenous spaces,Soviet. Math Dokl. 12 (1971), 1277–1491.MathSciNetGoogle Scholar
  32. [Mi3]
    V. D. Milman, The heritage of P. Lévy in geometrical functional analysis,Astérisque 157/158 (1988), 273–301.MathSciNetGoogle Scholar
  33. [P]
    G. Pisier, Probabilistic methods in the geometry of Banach spaces. Probability and Analysis, Varena (Italy) 1985,Lecture Notes in Math. 1206, Springer Verlag, 1986, 167–241.MathSciNetGoogle Scholar
  34. [R1]
    W. Rhee, On the fluctuations of the stochastic traveling salesperson problem,Math. of Operation Research 13 (1991), 482–489.MathSciNetCrossRefGoogle Scholar
  35. [R2]
    W. Rhee, A matching problem and subadditive Euclidean functionals,Ann. Applied Probab. 3 (1993), 794–801.MATHMathSciNetGoogle Scholar
  36. [R3]
    W. Rhee, On the fluctuations of simple matching,Oper. Res. Letters 16 (1994), 27–32.MATHCrossRefMathSciNetGoogle Scholar
  37. [R4]
    W. Rhee, Inequalities for the Bin Packing Problem III,Optimization 29 (1994), 381–385.MATHCrossRefMathSciNetGoogle Scholar
  38. [Ro]
    J. Rosinski, Remarks on a Strong Exponential Integrability of Vector Valued Random Series and Triangular Arrays,Ann. Probab., to appear.Google Scholar
  39. [R-T]
    W. Rhee, M. Talagrand, A sharp deviation inequality for the stochastic traveling salesman problemAnn. Probab. 17 (1989), 1–8.MATHMathSciNetGoogle Scholar
  40. [S]
    G. Schechtman, Levy type inequality for a class of metric spaces, Martingale Theory in Harmonic analysis and Banach spaces, Cleveland 1981,Lecture Note in Math. 939, Springer Verlag, 1981, 211–215.MathSciNetGoogle Scholar
  41. [S-S]
    E. Shamir, J. Spencer, Sharp concentration of the chromatic number of random graphs Gn, p,Combinatorica 7 (1987), 121–129.MATHCrossRefMathSciNetGoogle Scholar
  42. [T1]
    M. Talagrand, An isoperimetric theorem on the cube and the Kintchine Kahane inequalities,Proc. Amer. Math. Soc. 104 (1988), 905–909.MATHCrossRefMathSciNetGoogle Scholar
  43. [T2]
    M. Talagrand, Isoperimetry and integrability of the sum of independent Banach space valued random variables,Ann. Probab. 17 (1989), 1546–1570.MATHMathSciNetGoogle Scholar
  44. [T3]
    M. Talagrand, A new isoperimetric inequality for product measure, and the tails of sums of independent random variables,Geometric and Functional Analysis 1 (1991), 211–223.MATHCrossRefMathSciNetGoogle Scholar
  45. [T4]
    M. Talagrand, A new isoperimetric inequality for product measure, and the concentration of measure phenomenon, Israel Seminar (GAFA),Lecture Notes in Math. 1469, Springer Verlag, 1991, 94–124.CrossRefMathSciNetGoogle Scholar
  46. [T5]
    M. Talagrand, Regularity of infinitely divisible processes,Ann. Probab. 21 (1993), 362–432.MATHMathSciNetGoogle Scholar
  47. [T6]
    M. Talagrand, Supremum of some canonical processes,Amer. J. Math. 116 (1994), 295–314.CrossRefMathSciNetGoogle Scholar
  48. [T7]
    M. Talagrand, New concentration inequalities, in preparation.Google Scholar
  49. [W]
    D. W. Walkup, On the expected value of a random assignment problem,SIAM J. Comput. 8 (1979), 440–442.MATHCrossRefMathSciNetGoogle Scholar
  50. [Y]
    V. V. Yurinskii, Exponential bounds for large deviations,Theor. Prob. Appl. 19 (1974), 154–155.CrossRefGoogle Scholar

Copyright information

© Publications Mathématiques de L’I.H.É.S. 1995

Authors and Affiliations

  • Michel Talagrand
    • 1
  1. 1.Equipe d’Analyse - Tour 48 UA au CNRS no 754Université Paris VIParis Cedex 05

Personalised recommendations