Adaptive Mixtures of Local Experts

Abstract

We present a new supervised learning procedure for systems composed of many separate networks, each of which learns to handle a subset of the complete set of training cases. The new procedure can be viewed either as a modular version of a multilayer supervised network, or as an associative version of competitive learning. It therefore provides a new link between these two apparently different approaches. We demonstrate that the learning procedure divides up a vowel discrimination task into appropriate subtasks, each of which can be solved by a very simple expert network.

Cite

Text

Jacobs et al. "Adaptive Mixtures of Local Experts." Neural Computation, 1991. doi:10.1162/NECO.1991.3.1.79

Markdown

[Jacobs et al. "Adaptive Mixtures of Local Experts." Neural Computation, 1991.](https://mlanthology.org/neco/1991/jacobs1991neco-adaptive/) doi:10.1162/NECO.1991.3.1.79

BibTeX

@article{jacobs1991neco-adaptive,
  title     = {{Adaptive Mixtures of Local Experts}},
  author    = {Jacobs, Robert A. and Jordan, Michael I. and Nowlan, Steven J. and Hinton, Geoffrey E.},
  journal   = {Neural Computation},
  year      = {1991},
  pages     = {79-87},
  doi       = {10.1162/NECO.1991.3.1.79},
  volume    = {3},
  url       = {https://mlanthology.org/neco/1991/jacobs1991neco-adaptive/}
}