Numerical investigation of the asymmetric SK-model with deterministic dynamics

@article{Pfenning1991NumericalIO,
  title={Numerical investigation of the asymmetric SK-model with deterministic dynamics},
  author={T. Pfenning and H. Rieger and M. Schreckenberg},
  journal={Journal De Physique I},
  year={1991},
  volume={1},
  pages={323-329}
}
The dynam1c8 of the asymmetric SK-model with binary couplings Jij = + I, which was introduced in the context of neural networks, is investigated numerically at T = 0 and with sequential updating (deterministic dynamics). The system size8 are varied up to 8192 neurons. We find a uni- versal algebraic decay of the two- timestep autocorrelation fiJnction with exponent 3/2 for all values of the asymmetry. Furthermore the results sugge8t a transition (e.g. in the remanent magnetization and the… Expand
Attractors in fully asymmetric neural networks
The statistical properties of the length of the cycles and of the weights of the attraction basins in fully asymmetric neural networks (i.e. with completely uncorrelated synapses) are computed in theExpand
Relaxation, closing probabilities and transition from oscillatory to chaotic attractors in asymmetric neural networks
Attractors in asymmetric neural networks with deterministic parallel dynamics present a `chaotic' regime at symmetry where the average length of the cycles increases exponentially with system size,Expand
Statistical mechanics of neural networks
TLDR
It is demonstrated that a network storing a large number of sequences of patterns using a Hebb-like rule can store approximately twice as many patterns as the network trained with the Hebb rule to store individual patterns. Expand
On the stability of the mean-field spin glass broken phase under non-Hamiltonian perturbations
We study the dynamics of the SK model modified by a small non-Hamiltonian perturbation. We study aging, and we find that on the timescales investigated by our numerical simulations it survives aExpand
Subtle dynamic behaviour of finite-size Sherrington-Kirkpatrick spin glasses with nonsymmetric couplings
The authors have studied numerically the parallel dynamics of nonsymmetric Sherrington-Kirkpatrick spin glasses, varying the degree of symmetry eta :=(Ji, kJk, i)/(Ji, k2) of the couplingExpand
Remanence effects for spin glasses with sequential dynamics: exact results
The remanent magnetization and the remanent energy are calculated exactly for the +or-J spin glass in one dimension with random field and on Cayley trees for sequential dynamics at zero temperature.Expand
Finite signal transmission times and synaptic memory in neural networks.
  • Krüger, Martienssen, Rischke
  • Computer Science, Medicine
  • Physical review. E, Statistical physics, plasmas, fluids, and related interdisciplinary topics
  • 1995
TLDR
A significant reduction of the system's random response and a considerable enhancement of the response to stimuli with frequencies proportional to the inverse delay or memory time scale are found. Expand
Alignment in the fully asymmetric SK model
The long-time limit of the alignment function (or remanent energy E=-( sigma h)/2) of the fully asymmetric SK model is investigated analytically for parallel (Ep), sequential (Es) andExpand

References

and KIRKPAT & cK S
  • Phys . Rev Len .