Finite size effects and loss of self-averageness in the relaxational dynamics of the spherical Sherrington-Kirkpatrick model
arXiv:2103.12654 · doi:10.1088/1742-5468/ac0900
Abstract
We revisit the gradient descent dynamics of the spherical Sherrington-Kirkpatrick () model with finite number of degrees of freedom. For fully random initial conditions we confirm that the relaxation takes place in three time regimes: a first algebraic one controlled by the decay of the eigenvalue distribution of the random exchange interaction matrix at its edge in the infinite size limit; a faster algebraic one determined by the distribution of the gap between the two extreme eigenvalues; and a final exponential one determined by the minimal gap sampled in the disorder average. We also analyse the finite size effects on the relaxation from initial states which are almost projected on the saddles of the potential energy landscape, and we show that for deviations scaling as from perfect alignment the system escapes the initial configuration in a time-scale scaling as after which the dynamics no longer "self-averages" with respect to the initial conditions. We prove these statements with a combination of analytic and numerical methods.
22 pages, 8 figures
References in corpus (5)
- Theoretical perspective on the glass transition and amorphous materials
- Activated Aging Dynamics and Effective Trap Model Description in the Random Energy Model
- The mean field infinite range p=3 spin glass: equilibrium landscape and correlation time scales
- Non-linear susceptibilities of spherical models
- Revisiting the Concept of Activation in Supercooled Liquids
Cited by in corpus (5)
- Nonreciprocal Spin-Glass Transition and Aging
- Finite-size scaling in the ageing dynamics of the Glauber-Ising model
- Finite-size relaxational dynamics of a spike random matrix spherical model
- Exact finite-size scaling of the maximum likelihood spectra in the quenched and annealed Sherrington-Kirkpatrick spin glass
- Generalised Gibbs Ensemble for spherically constrained harmonic models