The action of external signals (constant, serrated, and its derivative, sinusoidal) on the RNN permit us to detect new specific features in the neural network dynamics, such as inertia, swapping, and depression. Besides phenomena such as flocking of active neurons, external stimulus amplification an
General properties of realistic neural network dynamics
β Scribed by K. Mardanov; Yu. Pis'mak; A. Potyagailo
- Publisher
- Elsevier Science
- Year
- 1997
- Tongue
- English
- Weight
- 415 KB
- Volume
- 34
- Category
- Article
- ISSN
- 0898-1221
No coin nor oath required. For personal study only.
β¦ Synopsis
The role of symmetry in the dynamics of the Realistic Neural Network model is studied. The phase diagram for Symmetric Realistic Neural Network with an external input of the constrained in time action is constructed. The oscillation regimes are investigated. For RNN of general form, the mean-field approximation is obtained.
π SIMILAR VOLUMES
The activation and weight dynamics of Artificial Neural Networks are derived from a partial differential equation (PDE) that may incorporate weights either as parameters or variables. It is shown that a single first-order Hamilton-Jacobi "parametricai" PDE suffices to derive the various neurodynamic
We consider a class of systems of differential equations in Nn which exhibits convergent dynamics. We find a Lyapunov function and show that every bounded trajectory converges to the set of equilibria. Our result generalizes the results of Cohen and Grossberg (1983) for convergent neural networks. I