ML Anthology

Rotor Neurons: Basic Formalism and Dynamics

Lars Gislén, Carsten Peterson, Bo Söderberg
NeCo 1992 pp. 737-745
doi:10.1162/NECO.1992.4.5.737 /neco/1992/gislen1992neco-rotor/

Abstract

Rotor neurons are introduced to encode states living on the surface of a sphere in D dimensions. Such rotors can be regarded as continuous generalizations of binary (Ising) neurons. The corresponding mean field equations are derived, and phase transition properties based on linearized dynamics are given. The power of this approach is illustrated with an optimization problem—placing N identical charges on a sphere such that the overall repulsive energy is minimized. The rotor approach appears superior to other methods for this problem both with respect to solution quality and computational effort needed.

NeCo Semantic Scholar

Cite

Text

Gislén et al. "Rotor Neurons: Basic Formalism and Dynamics." Neural Computation, 1992. doi:10.1162/NECO.1992.4.5.737

Markdown

[Gislén et al. "Rotor Neurons: Basic Formalism and Dynamics." Neural Computation, 1992.](https://mlanthology.org/neco/1992/gislen1992neco-rotor/) doi:10.1162/NECO.1992.4.5.737

BibTeX

@article{gislen1992neco-rotor,
  title     = {{Rotor Neurons: Basic Formalism and Dynamics}},
  author    = {Gislén, Lars and Peterson, Carsten and Söderberg, Bo},
  journal   = {Neural Computation},
  year      = {1992},
  pages     = {737-745},
  doi       = {10.1162/NECO.1992.4.5.737},
  volume    = {4},
  url       = {https://mlanthology.org/neco/1992/gislen1992neco-rotor/}
}