Back to Search
Start Over
IMPLEMENTING GAUSSIAN PROCESS INFERENCE WITH NEURAL NETWORKS.
- Source :
- International Journal of Neural Systems; Oct2006, Vol. 16 Issue 5, p321-327, 7p, 3 Diagrams, 2 Graphs
- Publication Year :
- 2006
-
Abstract
- Gaussian processes compare favourably with backpropagation neural networks as a tool for regression, and Bayesian neural networks have Gaussian process behaviour when the number of hidden neurons tends to infinity. We describe a simple recurrent neural network with connection weights trained by one-shot Hebbian learning. This network amounts to a dynamical system which relaxes to a stable state in which it generates predictions identical to those of Gaussian process regression. In effect an infinite number of hidden units in a feed-forward architecture can be replaced by a merely finite number, together with recurrent connections. [ABSTRACT FROM AUTHOR]
Details
- Language :
- English
- ISSN :
- 01290657
- Volume :
- 16
- Issue :
- 5
- Database :
- Complementary Index
- Journal :
- International Journal of Neural Systems
- Publication Type :
- Academic Journal
- Accession number :
- 23136432
- Full Text :
- https://doi.org/10.1142/S012906570600072X