Abstract
In order to develop effective evolutionary artificial neural networks (EANNs) we have to address the questions on how to evolve EANNs more efficiently and how to achieve the best performance from the ANNs evolved. Most of the previous works, however, do not utilize all the information obtained with several ANNs but choose the one best network in the last generation. Some recent works indicate that making use of population information by combining ANNs in the last generation can improve the performance, because they can complement each other to construct effective multiple neural networks. In this paper, we propose a new method of evolving multiple speciated neural networks by fitness sharing which helps to optimize multi-objective functions with genetic algorithms. Experiments with the breast cancer data from UCI benchmark datasets show that the proposed method can produce more speciated ANNs and improve the performance by combining the only representative individuals.
| Original language | English |
|---|---|
| Pages | 390-396 |
| Number of pages | 7 |
| Publication status | Published - 2001 Jan 1 |
| Event | Congress on Evolutionary Computation 2001 - Soul, Korea, Republic of Duration: 2001 May 27 → 2001 May 30 |
Other
| Other | Congress on Evolutionary Computation 2001 |
|---|---|
| Country | Korea, Republic of |
| City | Soul |
| Period | 01/5/27 → 01/5/30 |
Fingerprint
All Science Journal Classification (ASJC) codes
- Computer Science(all)
- Engineering(all)
Cite this
}
Speciated neural networks evolved with fitness sharing technique. / Ahn, J. H.; Cho, S. B.
2001. 390-396 Paper presented at Congress on Evolutionary Computation 2001, Soul, Korea, Republic of.Research output: Contribution to conference › Paper
TY - CONF
T1 - Speciated neural networks evolved with fitness sharing technique
AU - Ahn, J. H.
AU - Cho, S. B.
PY - 2001/1/1
Y1 - 2001/1/1
N2 - In order to develop effective evolutionary artificial neural networks (EANNs) we have to address the questions on how to evolve EANNs more efficiently and how to achieve the best performance from the ANNs evolved. Most of the previous works, however, do not utilize all the information obtained with several ANNs but choose the one best network in the last generation. Some recent works indicate that making use of population information by combining ANNs in the last generation can improve the performance, because they can complement each other to construct effective multiple neural networks. In this paper, we propose a new method of evolving multiple speciated neural networks by fitness sharing which helps to optimize multi-objective functions with genetic algorithms. Experiments with the breast cancer data from UCI benchmark datasets show that the proposed method can produce more speciated ANNs and improve the performance by combining the only representative individuals.
AB - In order to develop effective evolutionary artificial neural networks (EANNs) we have to address the questions on how to evolve EANNs more efficiently and how to achieve the best performance from the ANNs evolved. Most of the previous works, however, do not utilize all the information obtained with several ANNs but choose the one best network in the last generation. Some recent works indicate that making use of population information by combining ANNs in the last generation can improve the performance, because they can complement each other to construct effective multiple neural networks. In this paper, we propose a new method of evolving multiple speciated neural networks by fitness sharing which helps to optimize multi-objective functions with genetic algorithms. Experiments with the breast cancer data from UCI benchmark datasets show that the proposed method can produce more speciated ANNs and improve the performance by combining the only representative individuals.
UR - http://www.scopus.com/inward/record.url?scp=0034869569&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=0034869569&partnerID=8YFLogxK
M3 - Paper
AN - SCOPUS:0034869569
SP - 390
EP - 396
ER -