Abstract
The power of “hidden” neurons, i.e. neurons that do not directly communicate with the outside world, to perform complex tasks has become apparent in the case of layered feed-forward networks (perceptrons), discussed in . On the other hand, the theoretical treatment of such neural networks with strongly asymmetric synapses is much more difficult than that of symmetrically connected networks, developed in . It is therefore tempting to combine these two concepts, i.e. to study neural networks exhibiting hidden neurons and symmetric synapses.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
Notes
In electrical engineering this type of operation of an electrical circuit is called clamping of the input and output voltage.
However, a VLSI chip has been designed [A188] as a hardware representation of a Boltzmann machine which employs naturally occurring electrical noise as a source of thermal fluctuations.
Author information
Authors and Affiliations
Rights and permissions
Copyright information
© 1995 Springer-Verlag Berlin Heidelberg
About this chapter
Cite this chapter
Müller, B., Reinhardt, J., Strickland, M.T. (1995). Symmetrical Networks with Hidden Neurons. In: Neural Networks. Physics of Neural Networks. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-57760-4_13
Download citation
DOI: https://doi.org/10.1007/978-3-642-57760-4_13
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-60207-1
Online ISBN: 978-3-642-57760-4
eBook Packages: Springer Book Archive