Quick search Find article
Quick search
Find article

A numerical approach to large deviations in continuous time

Vivien Lecomte1,2 and Julien Tailleur3

Show affiliations


Figure 1

Figure 1. A history of the system between time t0 = 0 and time t.



Figure 2

Figure 2. Numerical evaluation of (1/LQ(s) for the Simple Exclusion Process (N = 200, L = 400). (a) Comparison between direct numerical measurement (blue crosses) and results from thermodynamic integration (red circles). (b) Comparison between numerical results (red circles) and analytical prediction (22) valid for small s (blue line).



Figure 3

Figure 3. Plot of the large deviation function (1/LQ(s) of the asymmetric simple exclusion process, for L = 400 sites and N = 200 particles. The jump rates are p = 1.2 and q = 0.8, whence E\simeq-0.2 . Blue crosses and red circles correspond to direct computation and thermodynamic integration, respectively. The asymmetry appears when comparing the extreme points s = ± 9.5.



Figure 4

Figure 4. (a) Average profile ρ for s = 0.3. To minimize the overall current, the system develops an asymmetric shock, where only the front particles can jump easily. (b) A typical configuration for s\gg
|E| . The particles are distributed almost uniformly. Note that |s| \gg |E| with s < 0 gives a similar result.



Figure 5

Figure 5. (a) Plot of the large deviation function (1/LK(s) associated to the number of events K in the contact process in a field (L = 120 sites). (b) The dynamical phase transition occurs at sc ~ 0.057. This is exemplified by plotting \psi_{K}'(s)=(1/ t)\langle K\rangle
_{s} for different system sizes (L = 4 in black, 8 in red, 15 in blue and 50 in magenta).




View by subject




Export








Please login to access our web services, or create an account if you don't yet have one.

You must have cookies enabled in your web browser to be able to login.

Username
Password

Forgotten your password? Get a new one here.