Attractor Neural Networks with Local Inhibition: From Statistical Physics to a Digitial Programmable Integrated Circuit

Abstract

Networks with local inhibition are shown to have enhanced compu(cid:173) tational performance with respect to the classical Hopfield-like net(cid:173) works. In particular the critical capacity of the network is increased as well as its capability to store correlated patterns. Chaotic dy(cid:173) namic behaviour (exponentially long transients) of the devices in(cid:173) dicates the overloading of the associative memory. An implementa(cid:173) tion based on a programmable logic device is here presented. A 16 neurons circuit is implemented whit a XILINK 4020 device. The peculiarity of this solution is the possibility to change parts of the project (weights, transfer function or the whole architecture) with a simple software download of the configuration into the XILINK chip.

Cite

Text

Pasero and Zecchina. "Attractor Neural Networks with Local Inhibition: From Statistical Physics to a Digitial Programmable Integrated Circuit." Neural Information Processing Systems, 1992.

Markdown

[Pasero and Zecchina. "Attractor Neural Networks with Local Inhibition: From Statistical Physics to a Digitial Programmable Integrated Circuit." Neural Information Processing Systems, 1992.](https://mlanthology.org/neurips/1992/pasero1992neurips-attractor/)

BibTeX

@inproceedings{pasero1992neurips-attractor,
  title     = {{Attractor Neural Networks with Local Inhibition: From Statistical Physics to a Digitial Programmable Integrated Circuit}},
  author    = {Pasero, E. and Zecchina, R.},
  booktitle = {Neural Information Processing Systems},
  year      = {1992},
  pages     = {805-812},
  url       = {https://mlanthology.org/neurips/1992/pasero1992neurips-attractor/}
}