Modern Analytic Techniques to Solve the Dynamics of Recurrent Neural Networks
Abstract
We describe the use of modern analytical techniques in solving the dynamics of symmetric and nonsymmetric recurrent neural net(cid:173) works near saturation. These explicitly take into account the cor(cid:173) relations between the post-synaptic potentials, and thereby allow for a reliable prediction of transients.
Cite
Text
Coolen et al. "Modern Analytic Techniques to Solve the Dynamics of Recurrent Neural Networks." Neural Information Processing Systems, 1995.Markdown
[Coolen et al. "Modern Analytic Techniques to Solve the Dynamics of Recurrent Neural Networks." Neural Information Processing Systems, 1995.](https://mlanthology.org/neurips/1995/coolen1995neurips-modern/)BibTeX
@inproceedings{coolen1995neurips-modern,
title = {{Modern Analytic Techniques to Solve the Dynamics of Recurrent Neural Networks}},
author = {Coolen, A.C.C. and Laughton, S. N. and Sherrington, D.},
booktitle = {Neural Information Processing Systems},
year = {1995},
pages = {253-259},
url = {https://mlanthology.org/neurips/1995/coolen1995neurips-modern/}
}