File Download
There are no files associated with this item.
Supplementary
-
Citations:
- Scopus: 0
- Appears in Collections:
Conference Paper: Some n-bit parity problems are solvable by feed-forward networks with less than n hidden units
| Title | Some n-bit parity problems are solvable by feed-forward networks with less than n hidden units |
|---|---|
| Authors | |
| Issue Date | 1993 |
| Citation | Proceedings Of The International Joint Conference On Neural Networks, 1993, v. 1, p. 305-308 How to Cite? |
| Abstract | Starting with two hidden units, we train a simple single hidden layer feed-forward neural network to solve the n-bit parity problem. If the network fails to recognize correctly all the input patterns, an additional hidden unit is added to the hidden layer and the network is retrained. This process is repeated until a network that correctly classifies all the input patterns has been constructed. Using a variant of the quasi-Newton methods for training, we have been able to find networks with a single layer containing less than n hidden units that solve the n-bit parity problem for some value of n. This proves the power of combining quasi-Newton method and node incremental approach. |
| Persistent Identifier | http://hdl.handle.net/10722/151800 |
| DC Field | Value | Language |
|---|---|---|
| dc.contributor.author | Setiono, Rudy | en_US |
| dc.contributor.author | Hui, Lucas Chi Kwong | en_US |
| dc.date.accessioned | 2012-06-26T06:29:43Z | - |
| dc.date.available | 2012-06-26T06:29:43Z | - |
| dc.date.issued | 1993 | en_US |
| dc.identifier.citation | Proceedings Of The International Joint Conference On Neural Networks, 1993, v. 1, p. 305-308 | en_US |
| dc.identifier.uri | http://hdl.handle.net/10722/151800 | - |
| dc.description.abstract | Starting with two hidden units, we train a simple single hidden layer feed-forward neural network to solve the n-bit parity problem. If the network fails to recognize correctly all the input patterns, an additional hidden unit is added to the hidden layer and the network is retrained. This process is repeated until a network that correctly classifies all the input patterns has been constructed. Using a variant of the quasi-Newton methods for training, we have been able to find networks with a single layer containing less than n hidden units that solve the n-bit parity problem for some value of n. This proves the power of combining quasi-Newton method and node incremental approach. | en_US |
| dc.language | eng | en_US |
| dc.relation.ispartof | Proceedings of the International Joint Conference on Neural Networks | en_US |
| dc.title | Some n-bit parity problems are solvable by feed-forward networks with less than n hidden units | en_US |
| dc.type | Conference_Paper | en_US |
| dc.identifier.email | Hui, Lucas Chi Kwong:hui@cs.hku.hk | en_US |
| dc.identifier.authority | Hui, Lucas Chi Kwong=rp00120 | en_US |
| dc.description.nature | link_to_subscribed_fulltext | en_US |
| dc.identifier.scopus | eid_2-s2.0-0027851228 | en_US |
| dc.identifier.volume | 1 | en_US |
| dc.identifier.spage | 305 | en_US |
| dc.identifier.epage | 308 | en_US |
| dc.identifier.scopusauthorid | Setiono, Rudy=7005033162 | en_US |
| dc.identifier.scopusauthorid | Hui, Lucas Chi Kwong=8905728300 | en_US |