Three-state neural network: from mutual information to the Hamiltonian.

@article{Dominguez2000ThreestateNN,
title={Three-state neural network: from mutual information to the Hamiltonian.},
author={D. R. Carreta Dominguez and E. Korutcheva},
journal={Physical review. E, Statistical physics, plasmas, fluids, and related interdisciplinary topics},
year={2000},
volume={62 2 Pt B},
pages={
2620-8
}
}

Physical review. E, Statistical physics, plasmas, fluids, and related interdisciplinary topics

The mutual information, I, of the three-state neural network can be obtained exactly for the mean-field architecture, as a function of three macroscopic parameters: the overlap, the neural activity and the activity-overlap, i.e., the overlap restricted to the active neurons. We perform an expansion of I on the overlap and the activity-overlap, around their values for neurons almost independent of the patterns. From this expansion we obtain an expression for a Hamiltonian which optimizes the… CONTINUE READING