Branching Brownian Motion Versus Random Energy Model in the Supercritical Phase: Overlap Distribution and Temperature Susceptibility

被引:0
作者
Bonnefont, Benjamin [1 ,2 ]
Pain, Michel [3 ]
Zindy, Olivier [1 ]
机构
[1] Sorbonne Paris Cite, Sorbonne Univ, CNRS, Lab Probabil Stat & Modelisat,UMR 8001, 4 Pl Jussieu, F-75252 Paris 05, France
[2] Univ Geneva, Sect Math, Rue Conseil Gen 7-9, CH-1205 Geneva, Switzerland
[3] Univ Toulouse, Inst Math Toulouse, CNRS, UMR 5219, 118 Route Narbonne, F-31062 Toulouse 9, France
关键词
Probability theory; Statistical physics; Branching Brownian motion; Random energy model; Gibbs measure; Overlap distribution; Temperature susceptibility; POISSON-DIRICHLET STATISTICS; EXTREME LEVEL SETS; EQUATION; FLUCTUATIONS; MAXIMUM; LAW; REM;
D O I
10.1007/s10955-025-03394-0
中图分类号
O4 [物理学];
学科分类号
0702 ;
摘要
In comparison with Derrida's REM, we investigate the influence of the so-called decoration processes arising in the limiting extremal processes of numerous log-correlated Gaussian fields. In particular, we focus on the branching Brownian motion and two specific quantities from statistical physics in the vicinity of the critical temperature. The first one is the two-temperature overlap, whose behavior at criticality is smoothened by the decoration process-unlike the one-temperature overlap which is identical-and the second one is the temperature susceptibility, as introduced by Sales and Bouchaud, which is strictly larger in the presence of decorations and diverges, close to the critical temperature, at the same speed as for the REM but with a different multiplicative constant. We also study some general decorated cases in order to highlight the fact that the BBM has a critical behavior in some sense to be made precise.
引用
收藏
页数:55
相关论文
共 53 条
  • [1] Branching Brownian motion seen from its tip
    Aidekon, E.
    Berestycki, J.
    Brunet, E.
    Shi, Z.
    [J]. PROBABILITY THEORY AND RELATED FIELDS, 2013, 157 (1-2) : 405 - 451
  • [2] Arguin L.-P., 2016, Extrema of Log-Correlated Random Variables: Principles and Examples, P166, DOI [10.1017/9781316403877.005, DOI 10.1017/9781316403877.005]
  • [3] Poisson-Dirichlet statistics for the extremes of the two-dimensional discrete Gaussian free field
    Arguin, Louis-Pierre
    Zindy, Olivier
    [J]. ELECTRONIC JOURNAL OF PROBABILITY, 2015, 20 : 1 - 19
  • [4] POISSON-DIRICHLET STATISTICS FOR THE EXTREMES OF A LOG-CORRELATED GAUSSIAN FIELD
    Arguin, Louis-Pierre
    Zindy, Olivier
    [J]. ANNALS OF APPLIED PROBABILITY, 2014, 24 (04) : 1446 - 1481
  • [5] The extremal process of branching Brownian motion
    Arguin, Louis-Pierre
    Bovier, Anton
    Kistler, Nicola
    [J]. PROBABILITY THEORY AND RELATED FIELDS, 2013, 157 (3-4) : 535 - 574
  • [6] Limit theorems for sums of random exponentials
    Ben Arous, G
    Bogachev, LV
    Molchanov, SA
    [J]. PROBABILITY THEORY AND RELATED FIELDS, 2005, 132 (04) : 579 - 612
  • [7] Biskup M, 2020, SPRINGER P MATH STAT, V304, P163, DOI 10.1007/978-3-030-32011-9_3
  • [8] Full extremal process, cluster law and freezing for the two-dimensional discrete Gaussian Free Field
    Biskup, Marek
    Louidor, Oren
    [J]. ADVANCES IN MATHEMATICS, 2018, 330 : 589 - 687
  • [9] Extreme Local Extrema of Two-Dimensional Discrete Gaussian Free Field
    Biskup, Marek
    Louidor, Oren
    [J]. COMMUNICATIONS IN MATHEMATICAL PHYSICS, 2016, 345 (01) : 271 - 304
  • [10] Bolthausen E, 2001, ANN PROBAB, V29, P1670