Please use this identifier to cite or link to this item:
https://doi.org/10.1088/0305-4470/28/1/005
| Title: | Synapse removal in discrete neural networks | Authors: | Viswanathan, R. | Issue Date: | 1995 | Citation: | Viswanathan, R. (1995). Synapse removal in discrete neural networks. Journal of Physics A: Mathematical and General 28 (1) : L25-L30. ScholarBank@NUS Repository. https://doi.org/10.1088/0305-4470/28/1/005 | Abstract: | The effect of removing synapses of particular magnitudes on the retrieval property of perceptron neural networks is determined by means of a space-of-interactions calculation. For a discrete state perceptron, an asymptotic (in number of input neurons) upper bound on the number of errors made in a set of given stored patterns due to synapse removal is computed as a function of the number and size of synapses removed. | Source Title: | Journal of Physics A: Mathematical and General | URI: | http://scholarbank.nus.edu.sg/handle/10635/111213 | ISSN: | 03054470 | DOI: | 10.1088/0305-4470/28/1/005 |
| Appears in Collections: | Staff Publications |
Show full item record
Files in This Item:
There are no files associated with this item.
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.