DMCA
Generalization in Interactive Networks: The Benefits of Inhibitory Competition and Hebbian Learning (2001)
| Venue: | Neural Computation |
| Citations: | 71 - 6 self |
BibTeX
@ARTICLE{O'reilly01generalizationin,
author = {Randall C. O'reilly},
title = {Generalization in Interactive Networks: The Benefits of Inhibitory Competition and Hebbian Learning},
journal = {Neural Computation},
year = {2001},
volume = {13},
pages = {1199--1242}
}
Share
 |
 |
 |
 |
||
OpenURL
Abstract
Computational models in cognitive neuroscience should ideally use biological properties and powerful computational principles to produce behavior consistent with psychological findings. Error-driven backpropagation is computationally powerful, and has proven useful for modeling a range of psychological data, but is not biologically plausible. Several approaches to implementing backpropagation in a biologically plausible fashion converge on the idea of using bidirectional activation propagation in interactive networks to convey error signals. This paper demonstrates two main points about these error-driven interactive networks: (a) they generalize poorly due to attractor dynamics that interfere with the network's ability to systematically produce novel combinatorial representations in response to novel inputs; and (b) this generalization problem can be remedied by adding two widely used mechanistic principles, inhibitory competition and Hebbian learning, that can be independent...
Keyphrases
inhibitory competition hebbian learning interactive network plausible fashion converge biological property psychological data psychological finding error-driven interactive network generalization problem error-driven backpropagation main point several approach powerful computational principle bidirectional activation propagation novel combinatorial representation error signal cognitive neuroscience computational model mechanistic principle
