Evaluation of Adaptive Mixtures of Competing Experts
Abstract
We compare the performance of the modular architecture, composed of competing expert networks, suggested by Jacobs, Jordan, Nowlan and Hinton (1991) to the performance of a single back-propagation network on a complex, but low-dimensional, vowel recognition task. Simulations reveal that this system is capable of uncovering interesting decompositions in a complex task. The type of decomposition is strongly influenced by the nature of the input to the gating network that decides which expert to use for each case. The modular architecture also exhibits consistently better generalization on many variations of the task.
Cite
Text
Nowlan and Hinton. "Evaluation of Adaptive Mixtures of Competing Experts." Neural Information Processing Systems, 1990.Markdown
[Nowlan and Hinton. "Evaluation of Adaptive Mixtures of Competing Experts." Neural Information Processing Systems, 1990.](https://mlanthology.org/neurips/1990/nowlan1990neurips-evaluation/)BibTeX
@inproceedings{nowlan1990neurips-evaluation,
title = {{Evaluation of Adaptive Mixtures of Competing Experts}},
author = {Nowlan, Steven J. and Hinton, Geoffrey E.},
booktitle = {Neural Information Processing Systems},
year = {1990},
pages = {774-780},
url = {https://mlanthology.org/neurips/1990/nowlan1990neurips-evaluation/}
}